Tag Archives: 90

Federate Database User Authentication Easily with IAM and Amazon Redshift

Post Syndicated from Thiyagarajan Arumugam original https://aws.amazon.com/blogs/big-data/federate-database-user-authentication-easily-with-iam-and-amazon-redshift/

Managing database users though federation allows you to manage authentication and authorization procedures centrally. Amazon Redshift now supports database authentication with IAM, enabling user authentication though enterprise federation. No need to manage separate database users and passwords to further ease the database administration. You can now manage users outside of AWS and authenticate them for access to an Amazon Redshift data warehouse. Do this by integrating IAM authentication and a third-party SAML-2.0 identity provider (IdP), such as AD FS, PingFederate, or Okta. In addition, database users can also be automatically created at their first login based on corporate permissions.

In this post, I demonstrate how you can extend the federation to enable single sign-on (SSO) to the Amazon Redshift data warehouse.

SAML and Amazon Redshift

AWS supports Security Assertion Markup Language (SAML) 2.0, which is an open standard for identity federation used by many IdPs. SAML enables federated SSO, which enables your users to sign in to the AWS Management Console. Users can also make programmatic calls to AWS API actions by using assertions from a SAML-compliant IdP. For example, if you use Microsoft Active Directory for corporate directories, you may be familiar with how Active Directory and AD FS work together to enable federation. For more information, see the Enabling Federation to AWS Using Windows Active Directory, AD FS, and SAML 2.0 AWS Security Blog post.

Amazon Redshift now provides the GetClusterCredentials API operation that allows you to generate temporary database user credentials for authentication. You can set up an IAM permissions policy that generates these credentials for connecting to Amazon Redshift. Extending the IAM authentication, you can configure the federation of AWS access though a SAML 2.0–compliant IdP. An IAM role can be configured to permit the federated users call the GetClusterCredentials action and generate temporary credentials to log in to Amazon Redshift databases. You can also set up policies to restrict access to Amazon Redshift clusters, databases, database user names, and user group.

Amazon Redshift federation workflow

In this post, I demonstrate how you can use a JDBC– or ODBC-based SQL client to log in to the Amazon Redshift cluster using this feature. The SQL clients used with Amazon Redshift JDBC or ODBC drivers automatically manage the process of calling the GetClusterCredentials action, retrieving the database user credentials, and establishing a connection to your Amazon Redshift database. You can also use your database application to programmatically call the GetClusterCredentials action, retrieve database user credentials, and connect to the database. I demonstrate these features using an example company to show how different database users accounts can be managed easily using federation.

The following diagram shows how the SSO process works:

  2. Authenticate using Corp Username/Password
  3. IdP sends SAML assertion
  4. Call STS to assume role with SAML
  5. STS Returns Temp Credentials
  6. Use Temp Credentials to get Temp cluster credentials
  7. Connect to Amazon Redshift using temp credentials


Example Corp. is using Active Directory (idp host:demo.examplecorp.com) to manage federated access for users in its organization. It has an AWS account: 123456789012 and currently manages an Amazon Redshift cluster with the cluster ID “examplecorp-dw”, database “analytics” in us-west-2 region for its Sales and Data Science teams. It wants the following access:

  • Sales users can access the examplecorp-dw cluster using the sales_grp database group
  • Sales users access examplecorp-dw through a JDBC-based SQL client
  • Sales users access examplecorp-dw through an ODBC connection, for their reporting tools
  • Data Science users access the examplecorp-dw cluster using the data_science_grp database group.
  • Partners access the examplecorp-dw cluster and query using the partner_grp database group.
  • Partners are not federated through Active Directory and are provided with separate IAM user credentials (with IAM user name examplecorpsalespartner).
  • Partners can connect to the examplecorp-dw cluster programmatically, using language such as Python.
  • All users are automatically created in Amazon Redshift when they log in for the first time.
  • (Optional) Internal users do not specify database user or group information in their connection string. It is automatically assigned.
  • Data warehouse users can use SSO for the Amazon Redshift data warehouse using the preceding permissions.

Step 1:  Set up IdPs and federation

The Enabling Federation to AWS Using Windows Active Directory post demonstrated how to prepare Active Directory and enable federation to AWS. Using those instructions, you can establish trust between your AWS account and the IdP and enable user access to AWS using SSO.  For more information, see Identity Providers and Federation.

For this walkthrough, assume that this company has already configured SSO to their AWS account: 123456789012 for their Active Directory domain demo.examplecorp.com. The Sales and Data Science teams are not required to specify database user and group information in the connection string. The connection string can be configured by adding SAML Attribute elements to your IdP. Configuring these optional attributes enables internal users to conveniently avoid providing the DbUser and DbGroup parameters when they log in to Amazon Redshift.

The user-name attribute can be set up as follows, with a user ID (for example, nancy) or an email address (for example. [email protected]):

<Attribute Name="https://redshift.amazon.com/SAML/Attributes/DbUser">  

The AutoCreate attribute can be defined as follows:

<Attribute Name="https://redshift.amazon.com/SAML/Attributes/AutoCreate">

The sales_grp database group can be included as follows:

<Attribute Name="https://redshift.amazon.com/SAML/Attributes/DbGroups">

For more information about attribute element configuration, see Configure SAML Assertions for Your IdP.

Step 2: Create IAM roles for access to the Amazon Redshift cluster

The next step is to create IAM policies with permissions to call GetClusterCredentials and provide authorization for Amazon Redshift resources. To grant a SQL client the ability to retrieve the cluster endpoint, region, and port automatically, include the redshift:DescribeClusters action with the Amazon Redshift cluster resource in the IAM role.  For example, users can connect to the Amazon Redshift cluster using a JDBC URL without the need to hardcode the Amazon Redshift endpoint:

Previous:  jdbc:redshift://endpoint:port/database

Current:  jdbc:redshift:iam://clustername:region/dbname

Use IAM to create the following policies. You can also use an existing user or role and assign these policies. For example, if you already created an IAM role for IdP access, you can attach the necessary policies to that role. Here is the policy created for sales users for this example:


    "Version": "2012-10-17",
    "Statement": [
            "Effect": "Allow",
            "Action": [
            "Resource": [
            "Effect": "Allow",
            "Action": [
            "Resource": [
            "Condition": {
                "StringEquals": {
                    "aws:userid": "AIDIODR4TAW7CSEXAMPLE:${redshift:DbUser}@examplecorp.com"
            "Effect": "Allow",
            "Action": [
            "Resource": [
            "Effect": "Allow",
            "Action": [
            "Resource": [

The policy uses the following parameter values:

  • Region: us-west-2
  • AWS Account: 123456789012
  • Cluster name: examplecorp-dw
  • Database group: sales_grp
Policy Statement Description

Allow users to retrieve the cluster endpoint, region, and port automatically for the Amazon Redshift cluster examplecorp-dw. This specification uses the resource format arn:aws:redshift:region:account-id:cluster:clustername. For example, the SQL client JDBC can be specified in the format jdbc:redshift:iam://clustername:region/dbname.

For more information, see Amazon Resource Names.


Generates a temporary token to authenticate into the examplecorp-dw cluster. “arn:aws:redshift:us-west-2:123456789012:dbuser:examplecorp-dw/${redshift:DbUser}” restricts the corporate user name to the database user name for that user. This resource is specified using the format: arn:aws:redshift:region:account-id:dbuser:clustername/dbusername.

The Condition block enforces that the AWS user ID should match “AIDIODR4TAW7CSEXAMPLE:${redshift:DbUser}@examplecorp.com”, so that individual users can authenticate only as themselves. The AIDIODR4TAW7CSEXAMPLE role has the Sales_DW_IAM_Policy policy attached.

Automatically creates database users in examplecorp-dw, when they log in for the first time. Subsequent logins reuse the existing database user.
Allows sales users to join the sales_grp database group through the resource “arn:aws:redshift:us-west-2:123456789012:dbgroup:examplecorp-dw/sales_grp” that is specified in the format arn:aws:redshift:region:account-id:dbgroup:clustername/dbgroupname.

Similar policies can be created for Data Science users with access to join the data_science_grp group in examplecorp-dw. You can now attach the Sales_DW_IAM_Policy policy to the role that is mapped to IdP application for SSO.
 For more information about how to define the claim rules, see Configuring SAML Assertions for the Authentication Response.

Because partners are not authorized using Active Directory, they are provided with IAM credentials and added to the partner_grp database group. The Partner_DW_IAM_Policy is attached to the IAM users for partners. The following policy allows partners to log in using the IAM user name as the database user name.


    "Version": "2012-10-17",
    "Statement": [
            "Effect": "Allow",
            "Action": [
            "Resource": [
            "Effect": "Allow",
            "Action": [
            "Resource": [
            "Condition": {
                "StringEquals": {
                    "redshift:DbUser": "${aws:username}"
            "Effect": "Allow",
            "Action": [
            "Resource": [
            "Effect": "Allow",
            "Action": [
            "Resource": [

redshift:DbUser“: “${aws:username}” forces an IAM user to use the IAM user name as the database user name.

With the previous steps configured, you can now establish the connection to Amazon Redshift through JDBC– or ODBC-supported clients.

Step 3: Set up database user access

Before you start connecting to Amazon Redshift using the SQL client, set up the database groups for appropriate data access. Log in to your Amazon Redshift database as superuser to create a database group, using CREATE GROUP.

Log in to examplecorp-dw/analytics as superuser and create the following groups and users:

CREATE GROUP sales_grp;
CREATE GROUP datascience_grp;
CREATE GROUP partner_grp;

Use the GRANT command to define access permissions to database objects (tables/views) for the preceding groups.

Step 4: Connect to Amazon Redshift using the JDBC SQL client

Assume that sales user “nancy” is using the SQL Workbench client and JDBC driver to log in to the Amazon Redshift data warehouse. The following steps help set up the client and establish the connection:

  1. Download the latest Amazon Redshift JDBC driver from the Configure a JDBC Connection page
  2. Build the JDBC URL with the IAM option in the following format:

Because the redshift:DescribeClusters action is assigned to the preceding IAM roles, it automatically resolves the cluster endpoints and the port. Otherwise, you can specify the endpoint and port information in the JDBC URL, as described in Configure a JDBC Connection.

Identify the following JDBC options for providing the IAM credentials (see the “Prepare your environment” section) and configure in the SQL Workbench Connection Profile:

idp_host=demo.examplecorp.com (The name of the corporate identity provider host)
idp_port=443  (The port of the corporate identity provider host)
user=examplecorp\nancy(corporate user name)
password=***(corporate user password)

The SQL workbench configuration looks similar to the following screenshot:

Now, “nancy” can connect to examplecorp-dw by authenticating using the corporate Active Directory. Because the SAML attributes elements are already configured for nancy, she logs in as database user nancy and is assigned the sales_grp. Similarly, other Sales and Data Science users can connect to the examplecorp-dw cluster. A custom Amazon Redshift ODBC driver can also be used to connect using a SQL client. For more information, see Configure an ODBC Connection.

Step 5: Connecting to Amazon Redshift using JDBC SQL Client and IAM Credentials

This optional step is necessary only when you want to enable users that are not authenticated with Active Directory. Partners are provided with IAM credentials that they can use to connect to the examplecorp-dw Amazon Redshift clusters. These IAM users are attached to Partner_DW_IAM_Policy that assigns them to be assigned to the public database group in Amazon Redshift. The following JDBC URLs enable them to connect to the Amazon Redshift cluster:

jdbc:redshift:iam//examplecorp-dw/analytics?AccessKeyID=XXX&SecretAccessKey=YYY&DbUser=examplecorpsalespartner&DbGroup= partner_grp&AutoCreate=true

The AutoCreate option automatically creates a new database user the first time the partner logs in. There are several other options available to conveniently specify the IAM user credentials. For more information, see Options for providing IAM credentials.

Step 6: Connecting to Amazon Redshift using an ODBC client for Microsoft Windows

Assume that another sales user “uma” is using an ODBC-based client to log in to the Amazon Redshift data warehouse using Example Corp Active Directory. The following steps help set up the ODBC client and establish the Amazon Redshift connection in a Microsoft Windows operating system connected to your corporate network:

  1. Download and install the latest Amazon Redshift ODBC driver.
  2. Create a system DSN entry.
    1. In the Start menu, locate the driver folder or folders:
      • Amazon Redshift ODBC Driver (32-bit)
      • Amazon Redshift ODBC Driver (64-bit)
      • If you installed both drivers, you have a folder for each driver.
    2. Choose ODBC Administrator, and then type your administrator credentials.
    3. To configure the driver for all users on the computer, choose System DSN. To configure the driver for your user account only, choose User DSN.
    4. Choose Add.
  3. Select the Amazon Redshift ODBC driver, and choose Finish. Configure the following attributes:
    Data Source Name =any friendly name to identify the ODBC connection 
    user=uma(corporate user name)
    Auth Type-Identity Provider: AD FS
    password=leave blank (Windows automatically authenticates)
    Cluster ID: examplecorp-dw
    idp_host=demo.examplecorp.com (The name of the corporate IdP host)

This configuration looks like the following:

  1. Choose OK to save the ODBC connection.
  2. Verify that uma is set up with the SAML attributes, as described in the “Set up IdPs and federation” section.

The user uma can now use this ODBC connection to establish the connection to the Amazon Redshift cluster using any ODBC-based tools or reporting tools such as Tableau. Internally, uma authenticates using the Sales_DW_IAM_Policy  IAM role and is assigned the sales_grp database group.

Step 7: Connecting to Amazon Redshift using Python and IAM credentials

To enable partners, connect to the examplecorp-dw cluster programmatically, using Python on a computer such as Amazon EC2 instance. Reuse the IAM users that are attached to the Partner_DW_IAM_Policy policy defined in Step 2.

The following steps show this set up on an EC2 instance:

  1. Launch a new EC2 instance with the Partner_DW_IAM_Policy role, as described in Using an IAM Role to Grant Permissions to Applications Running on Amazon EC2 Instances. Alternatively, you can attach an existing IAM role to an EC2 instance.
  2. This example uses Python PostgreSQL Driver (PyGreSQL) to connect to your Amazon Redshift clusters. To install PyGreSQL on Amazon Linux, use the following command as the ec2-user:
    sudo easy_install pip
    sudo yum install postgresql postgresql-devel gcc python-devel
    sudo pip install PyGreSQL

  1. The following code snippet demonstrates programmatic access to Amazon Redshift for partner users:
    #!/usr/bin/env python
    python redshift-unload-copy.py <config file> <region>
    * Copyright 2014, Amazon.com, Inc. or its affiliates. All Rights Reserved.
    * Licensed under the Amazon Software License (the "License").
    * You may not use this file except in compliance with the License.
    * A copy of the License is located at
    * http://aws.amazon.com/asl/
    * or in the "license" file accompanying this file. This file is distributed
    * express or implied. See the License for the specific language governing
    * permissions and limitations under the License.
    import sys
    import pg
    import boto3
    REGION = 'us-west-2'
    CLUSTER_IDENTIFIER = 'examplecorp-dw'
    DB_NAME = 'sales_db'
    DB_USER = 'examplecorpsalespartner'
    options = """keepalives=1 keepalives_idle=200 keepalives_interval=200
    set_timeout_stmt = "set statement_timeout = 1200000"
    def conn_to_rs(host, port, db, usr, pwd, opt=options, timeout=set_timeout_stmt):
        rs_conn_string = """host=%s port=%s dbname=%s user=%s password=%s
                             %s""" % (host, port, db, usr, pwd, opt)
        print "Connecting to %s:%s:%s as %s" % (host, port, db, usr)
        rs_conn = pg.connect(dbname=rs_conn_string)
        return rs_conn
    def main():
        # describe the cluster and fetch the IAM temporary credentials
        global redshift_client
        redshift_client = boto3.client('redshift', region_name=REGION)
        response_cluster_details = redshift_client.describe_clusters(ClusterIdentifier=CLUSTER_IDENTIFIER)
        response_credentials = redshift_client.get_cluster_credentials(DbUser=DB_USER,DbName=DB_NAME,ClusterIdentifier=CLUSTER_IDENTIFIER,DurationSeconds=3600)
        rs_host = response_cluster_details['Clusters'][0]['Endpoint']['Address']
        rs_port = response_cluster_details['Clusters'][0]['Endpoint']['Port']
        rs_db = DB_NAME
        rs_iam_user = response_credentials['DbUser']
        rs_iam_pwd = response_credentials['DbPassword']
        # connect to the Amazon Redshift cluster
        conn = conn_to_rs(rs_host, rs_port, rs_db, rs_iam_user,rs_iam_pwd)
        # execute a query
        result = conn.query("SELECT sysdate as dt")
        # fetch results from the query
        for dt_val in result.getresult() :
            print dt_val
        # close the Amazon Redshift connection
    if __name__ == "__main__":

You can save this Python program in a file (redshiftscript.py) and execute it at the command line as ec2-user:

python redshiftscript.py

Now partners can connect to the Amazon Redshift cluster using the Python script, and authentication is federated through the IAM user.


In this post, I demonstrated how to use federated access using Active Directory and IAM roles to enable single sign-on to an Amazon Redshift cluster. I also showed how partners outside an organization can be managed easily using IAM credentials.  Using the GetClusterCredentials API action, now supported by Amazon Redshift, lets you manage a large number of database users and have them use corporate credentials to log in. You don’t have to maintain separate database user accounts.

Although this post demonstrated the integration of IAM with AD FS and Active Directory, you can replicate this solution across with your choice of SAML 2.0 third-party identity providers (IdP), such as PingFederate or Okta. For the different supported federation options, see Configure SAML Assertions for Your IdP.

If you have questions or suggestions, please comment below.

Additional Reading

Learn how to establish federated access to your AWS resources by using Active Directory user attributes.

About the Author

Thiyagarajan Arumugam is a Big Data Solutions Architect at Amazon Web Services and designs customer architectures to process data at scale. Prior to AWS, he built data warehouse solutions at Amazon.com. In his free time, he enjoys all outdoor sports and practices the Indian classical drum mridangam.



Post Syndicated from Антония original http://dni.li/2017/10/19/redtape/

Подавам някаква молба в общината.

Първо ходене

Дават ми бланка и ми казват какви документи трябва да нося.

Второ ходене

Връщам попълнената бланка и предоставям всички документи (ОСЕМ на брой), които са изискали от мен – в оригинал и с по едно копие. Оригиналът бил да го покажа само, а копията ги завеждат в „досие“. Всичко уж е ОК. Но седмица по-късно…

Първо обаждане по телефона

Жизнерадостна леля ме уведомява, че от нейния отдел на общината ми искат документ за идентичност на имената. Въпросният документ се издава от съседния отдел на общината. Но трябва аз да си го поискам, те вътрешнообщински не можело да общуват.

Трето ходене в общината

Нося документа. И ме светват, че трябва още нещо да свърша – да извадя някакво удостоверение от на-майна-си-райна организация (НМСРО) и да го предоставя в 14-дневен срок.

Първо ходене в НМСРО

Учтиво, но твърдо ме уверяват, че въпросното удостоверение се издава в тримесечен срок. И дори и да се напънат – не зависело от тях, имало и странични фактори, нааай-бързо евентуално за 2 месеца щели да се справят. Тук губя нерви, разбира се, и почвам да крещя. Спокойно ми отвръщат, че и да крещя, и да не крещя… Защо не се срещна с шефката на НМСРО. Която повтаря думите на персонала си и вдига рамене, няма какво да направи. Защо съм се ядосвала на подобни неразбории, то тук е така. И всъщност тя не била сигурни, че общината трябвало да изисква подобен документ от тях.

Второ обаждане в общината

Този път аз ги набирам и им казвам, че съм в кабинета на шефката на НМСРО, която първо ми е казала, че въпросното удостоверение ще е готово след 3 месеца и няма как да спазя двуседмичния срок, който общината ми е дала, и второ ме е светнала, че WTF ми искат подобни неща да вадя. Лелката не е много жизнерадостна вече, то от нея нищо не зависело, нейната шефка ѝ била казала да направи това. Защо не отида отново в общината да говоря директно с висшестоящите.

Четвърто ходене в общината

Шефката на отдела е благосклонна – ще „задвижи моя въпрос“, но все пак трябвало да ѝ занеса удостоверението, пък било то и след три месеца.

Та така. В началото на ноември ще разберем дали девет подадени документа и един в процес на изваждане, четири разкарвания до общината, едно ходене до НМСРО и два телефонни разговора ще са достатъчни, за да си свърши някой работата.

Говорилнята около @tourbg

Post Syndicated from Боян Юруков original https://yurukov.net/blog/2017/tourbg/

Изминаха 10 дни откакто започна да се говори за Александър Николов/tourbg/Спас и какво е правил. Изявиха се доста анализатори с претенции, че имат пръст на пулса на социалните медии, модерното общество, „умните и красивите“, „новата буржоазия“ и прочие епитети. Скроиха се схеми, превърнаха ония в жертва и герой на „обикновения човек“, посрамиха го после, посрамиха жертвите му, оправдаха го, оправдаха полицията и всичко това още продължава. Сагата се превърна повече е нарицателно, отколкото в казус и затова нямам намерение да я коментирам тук.

Вместо това реших да направя друго. Подобно на няколко други бури като #siromahovfacts и #toplomovies свалих цялата активност в Twitter и ще ви покажа кога и колко е говорено за това.

По ключови думи

Търсил съм по няколко термина видими долу. При „спас“ включих само tweet-овете, които са маркирани от Twitter, че са на български. Думата се използва доста в руски и сръбски съобщения. При „билети“ и „спас“ несъмнено има няколко, които не са свързани, но съдейки по активността преди 7-ми, те са единици. Забелязват се пиковете около обявяването на новини около случая.

Най-активно пишещи

Най-активни са @varnasummer и @NewsMixerBG, а след тях с над 3 пъти по-ниска активност са @Tangerrinka и @nervnata. Всъщност, почти всичко от @varnasummer е на 9-ти около обяд.

Spinrilla Wants RIAA Case Thrown Out Over ‘Lies’ About ‘Hidden’ Piracy Data

Post Syndicated from Ernesto original https://torrentfreak.com/spinrilla-wants-riaa-case-thrown-out-over-lies-about-hidden-piracy-data-171016/

Earlier this year, a group of well-known labels targeted Spinrilla, a popular hip-hop mixtape site and app which serves millions of users.

The coalition of record labels, including Sony Music, Warner Bros. Records, and Universal Music Group, filed a lawsuit against the service over alleged copyright infringements.

While the discovery process is still ongoing, Spinrilla recently informed the court that the record labels have “just about derailed” the entire case. The company has submitted a motion for sanctions, which is currently sealed, but additional information submitted to the court this week reveals what’s going on.

When the labels filed their original complaint they listed 210 tracks, without providing the allegedly infringing URLs. These weren’t shared during the early stages of the discovery process either, forcing the site to manually search for potentially infringing links.

Then, early October, Spinrilla received a massive spreadsheet with over 2,000 tracks, including the infringing URLs. This data came from the RIAA and supported the long list of infringements in the amended complaint submitted around the same time.

The spreadsheet would have made the discovery process much easier for Spinrilla. In a supplemental brief supporting a motion for sanctions, Spinrilla accuses the labels of hiding the piracy data from them and lying about it, “derailing” the case in the process.

“Significantly, Plaintiffs used that lie to convince the Court they should be allowed to add about 1,900 allegedly infringed sound recordings to their original list of 210. Later, Plaintiffs repeated that lie to convince the Court to give them time to add even more sound recordings to their list.”


Spinrilla says they were forced to go down an expensive and unnecessary rabbit hole to find the infringing files, even though the RIAA data was available all along.

“By hiding and lying about the RIAA data, Plaintiffs forced Defendants to spend precious time and money fumbling through discovery. Not knowing that Plaintiffs had the RIAA data,” the company writes.

The hip-hop mixtape site argues that the alleged wrongdoing is severe enough to have the entire complaint dismissed, as the ultimate sanction.

“It is without exaggeration to say that by hiding the RIAA spreadsheets and that underlying data, Defendants have been severely prejudiced. The Complaint should be dismissed with prejudice and, if it is, Plaintiffs can only blame themselves,” Spinrilla concludes.

The stakes are certainly high in this case. With well over 2,000 infringing tracks listed in the amended complaint, the hip-hop mixtape site faces statutory damages as high as $300 million, at least in theory.

Spinrilla’s supplement brief in further support of the motion for sanctions is available here (pdf).

Source: TF, for the latest info on copyright, file-sharing, torrent sites and ANONYMOUS VPN services.

За един Иван и измислените барикади

Post Syndicated from Bozho original https://blog.bozho.net/blog/2976

И аз като Венци Мицов имам един приятел Иван. Дето живее „в провинцията“ (откъдето съм и аз). И който взема, айде не 450 лева, ама твърде малко пари, за да изхранва семейството си.

Иван работи доста за тези пари – не т.нар. работа на хора с бели якички, но работата си е работа.

Иван не го интересува кой е на власт, щото всичките са едни и същи и той не вижда много много разлика. Не слуша вече и предизборния студия, не чете секцията за политика във вестника, защото му е писнало.

Не го интересуват стратегии и концепции. И не защото не е учил „стратегическо планиране“, а защото тия стратегии и концепции в обозримо бъдеще няма да му вдигнат значително заплатата, няма да го направят по-здрав (щото здравеопазването уж е безплатно, ама навсякъде плащаш), няма да го направят и по-щастлив.

Само че се появяват едни хора, които решават да превърнат Иван в образ. Да го поставят като контрапункт на хора, които не харесват. Да създадат едно измислено разделение. И после да могат да кажат на Иван „Иване, виж ги тия как са се ояли и говорят глупости. Маскари. Ама ти мене ше слушаш“.

Тия хора заливат Иван с това разделение по всички канали, с които разполагат. Обясняват му как бюрократите в Европа вземат милиони и само тормозят хора като него. Обясняват му как щом не може да си сложи нещо на масата или да го увие в амбалажна хартия, то сигурно е някаква глупост.

Рисуват картина на две Българии, едната на Иван, а другата на някакъв имагинерен елит.

Не че няма градски хора, за които важната тема на деня е дали небостъргачът в другия край на София (в който не са стъпвали от години) ще кореспондира със стила на околното строителство, а не дали хора като Иван могат да се оправят в живота. Обаче то си е тяхно право да имат собствени приоритети, стига да не налагат тези приоритети на всички. И колкото и фейсбук да създава такова впечатление, те едва ли го правят.

А колкото до стратегиите и визиите – те са безсмислени ако няма кой да ги изпълни. Накрая ще ги изпълняват хора като Иван, най-вероятно.

Обаче опитът да се обясни на Иван, че това всичко са глупости, създава у Иван един нихилизъм. На него му е все повече все тая, все повече не смята, че нещо може да се промени и все повече цветовете се сливат в два-три нюанса на сивото. А като чуе нещо позитивно по телевизията, инстинктивно решава, че „пак ни лъжат“.

Да, стратегиите няма да му вдигнат заплатата утре, визиите няма да му сложат по-качествена храна на масата, а концепциите няма да му оправят течащия покрив. Обаче кое ще?

Иван не е част от друга България, а различията в социалните възможности, макар и напълно реални, не са това, заради което Иван е беден. Иван е беден, защото едни хора го използват, за да бъдат „елит за един ден“, за да източат де що има да се източва, да не пускат без рушвет чужди компании, които биха давали на Иван повече пари. Но тези хора обясняват на Иван, че е беден, защото стотина човека в центъра на София не мислели за него, защото Брюксел регулирал краставиците и защото сме си прецакали отношенията с „братска Русия“, където сме изнасяли стоки за милиони!

Словесното построяване на такива разделения ги превръща в реалност, обаче. Хора застават от двете страни на тези имагинерни барикади. Хора, които не биха имали проблем да седнат на една маса и да намерят общи теми в иначе разнородните си ежедневия.

А в сенките отстрани на тези барикади едни хитри хора потриват ръце.

Pirate Bay’s Iconic .SE Domain has Expired (Updated)

Post Syndicated from Ernesto original https://torrentfreak.com/pirate-bays-iconic-se-domain-has-expired-and-is-for-sale-171016/

When The Pirate Bay first came online during the summer of 2003, its main point of access was thepiratebay.org.

Since then the site has burnt through more than a dozen domains, trying to evade seizures or other legal threats.

For many years thepiratebay.se operated as the site’s main domain name. Earlier this year the site moved back to the good old .org again, and from the looks of it, TPB is ready to say farewell to the Swedish domain.

Thepiratebay.se expired last week and, if nothing happens, it will be de-activated tomorrow. This means that the site might lose control over a piece of its history.

The torrent site moved from the ORG to the SE domain in 2012, fearing that US authorities would seize the former. Around that time the Department of Homeland Security took hundreds of sites offline and the Pirate Bay team feared that they would be next.

Thepiratebay.se has expired

Ironically, however, the next big threat came from Sweden, the Scandinavian country where the site once started.

In 2013, a local anti-piracy group filed a motion targeting two of The Pirate Bay’s domains, ThePirateBay.se and PirateBay.se. This case that has been dragging on for years now.

During this time TPB moved back and forth between domains but the .se domain turned out to be a safer haven than most alternatives, despite the legal issues. Many other domains were simply seized or suspended without prior notice.

When the Swedish Court of Appeal eventually ruled that The Pirate Bay’s domain had to be confiscated and forfeited to the state, the site’s operators moved back to the .org domain, where it all started.

Although a Supreme Court appeal is still pending, according to a report from IDG earlier this year the court has placed a lock on the domain. This prevents the owner from changing or transferring it, which may explain why it has expired.

The lock is relevant, as the domain not only expired but has also been put of for sale again in the SEDO marketplace, with a minimum bid of $90. This sale would be impossible, if the domain is locked.

Thepiratebay.se for sale

Perhaps the most ironic of all is the fact that TPB moved to .se because it feared that the US controlled .org domain was easy prey.

Fast forward half a decade and over a dozen domains have come and gone while thepiratebay.org still stands strong, despite entertainment industry pressure.

Update: We updated the article to mention that the domain name is locked by the Swedish Supreme Court. This means that it can’t be updated and would explain why it has expired.

Source: TF, for the latest info on copyright, file-sharing, torrent sites and ANONYMOUS VPN services.

Господа министри, мерете си данните

Post Syndicated from Bozho original https://blog.bozho.net/blog/2965

Когато някой „по телевизора“ изкаже някакво твърдение, никога не е ясно откъде са му данните. Разпространяват се доста митове, основани на гледане в тавана. Но в тези случаи има поне частично обяснение – хората може би просто няма откъде да вземат данните, за да ги анализират.

Не така стоят нещата с министрите, обаче. Министрите (и председателите на агенции) разполагат с администрация, която може да им даде данни. Прави впечатление обаче фриволното боравене с метарията, и разпространяване на данни, които просто са грешни. Боян Юруков вече е писал за председателя на агенцията за българите в чужбина, който напълно необосновано обяви, че има 6-7 милиона българи в чужбина.

Аз ще се спра на министъра на околната среда, Нено Димов, който през седмицата е обявил, че „през 2016 г. в столицата е имало едва две минимални превишения на показателите за замърсяване с фини прахови частици“. Това ми звучеше доста малко вероятно, предвид, че данните за 2015-та, които съм разглеждал, показваха 60 дни над нормата (при допустими 35). За 1 година такъв напредък, въпреки позитивния тренд, изглеждаше невероятен.

За съжаление ИАОС не публикува суровите данни във формата, в който бях взел тези до 2015 (с искане за обществена информация), но благодарение на инициативата за отворени данни все пак всеки ден качват данни от бюлетина за качеството на въздуха, където пише коя станция е превишила дадена норма. На база на тези данни не мога да направя същите графики като в предишния анализ, но мога да проверя твърдението на министъра.

И то, разбира се, се оказа грешно. Ето броя дни, в които нормата е превишена:

Превишена стойност в поне 1 станция: 74 дни
Превишена стойност в поне 2 станции: 48 дни
Превишена стойност в поне 3 станции: 43 дни
Превишена стойност в поне 4 станции: 36 дни
Превишена стойност в поне 5 станции: 15 дни

По спомен, станциите са 6, но последната е на Копитото и там винаги е чисто. С наличните данни (които са качени за 340, а не за 365 дни) не мога да кажа за средната стойност за града, но когато 4 от 5 станции имат превишение 36 дни (1 над европейската норма), министърът просто изнася грешни данни. Или „е имал предвид друго“, в който случай – нека обясни.

Пак подчертавам, че трендът наистина изглежда позитивен. Също така приветствам вземането на мерки срещу замърсяването от страна на министерството – именно национална политика по въпроса е начинът за решаване на проблема. Но не е редно да имаш цялата администрация на МОСВ и ИАОС под себе си и да кажеш, че само 2 дни била превишена нормата. Надявам се мерките, които се подготвят, да са основание на по-верни данни.

Всъщност, трендът може би изглежда възходящ заради факта, че от няколко години измервателната станция на Орлов мост е премахната. Там, разбира се, фините прахови частици са в най-големи количества. Вероятно има разумно обяснение за премахването (наистина Орлов мост е граничен случай), но трябва да го имаме предвид.

И в това всъщност е част от проблема – в София има доста малко измервателни станции, за да придобием пълна картина. Най-близката станция до мен е на километри. За щастие има проекта airbg.info, чрез който всеки може да си постави измервателна станция и да докладва данните. Така се създава доста по-пълна картина на замърсяването. В съботната сутрин, без мъгли и без нужда от сериозно отопление, картата на София изглежда добре.

Но да се върнем на министрите и данните. Политиката за отворени данни има за цел както повече прозрачност, така и по-информирани решения в управлението. Второто засега не изглежда да е постигнато, решения продължават често да се вземат „по интуиция“, а официални лица продължават да разпространяват неосновани на данни твърдения. Но поне данните ги има, та гражданите можем да посочим грешките.

Predict Billboard Top 10 Hits Using RStudio, H2O and Amazon Athena

Post Syndicated from Gopal Wunnava original https://aws.amazon.com/blogs/big-data/predict-billboard-top-10-hits-using-rstudio-h2o-and-amazon-athena/

Success in the popular music industry is typically measured in terms of the number of Top 10 hits artists have to their credit. The music industry is a highly competitive multi-billion dollar business, and record labels incur various costs in exchange for a percentage of the profits from sales and concert tickets.

Predicting the success of an artist’s release in the popular music industry can be difficult. One release may be extremely popular, resulting in widespread play on TV, radio and social media, while another single may turn out quite unpopular, and therefore unprofitable. Record labels need to be selective in their decision making, and predictive analytics can help them with decision making around the type of songs and artists they need to promote.

In this walkthrough, you leverage H2O.ai, Amazon Athena, and RStudio to make predictions on whether a song might make it to the Top 10 Billboard charts. You explore the GLM, GBM, and deep learning modeling techniques using H2O’s rapid, distributed and easy-to-use open source parallel processing engine. RStudio is a popular IDE, licensed either commercially or under AGPLv3, for working with R. This is ideal if you don’t want to connect to a server via SSH and use code editors such as vi to do analytics. RStudio is available in a desktop version, or a server version that allows you to access R via a web browser. RStudio’s Notebooks feature is used to demonstrate the execution of code and output. In addition, this post showcases how you can leverage Athena for query and interactive analysis during the modeling phase. A working knowledge of statistics and machine learning would be helpful to interpret the analysis being performed in this post.


Your goal is to predict whether a song will make it to the Top 10 Billboard charts. For this purpose, you will be using multiple modeling techniques―namely GLM, GBM and deep learning―and choose the model that is the best fit.

This solution involves the following steps:

  • Install and configure RStudio with Athena
  • Log in to RStudio
  • Install R packages
  • Connect to Athena
  • Create a dataset
  • Create models

Install and configure RStudio with Athena

Use the following AWS CloudFormation stack to install, configure, and connect RStudio on an Amazon EC2 instance with Athena.

Launching this stack creates all required resources and prerequisites:

  • Amazon EC2 instance with Amazon Linux (minimum size of t2.large is recommended)
  • Provisioning of the EC2 instance in an existing VPC and public subnet
  • Installation of Java 8
  • Assignment of an IAM role to the EC2 instance with the required permissions for accessing Athena and Amazon S3
  • Security group allowing access to the RStudio and SSH ports from the internet (I recommend restricting access to these ports)
  • S3 staging bucket required for Athena (referenced within RStudio as ATHENABUCKET)
  • RStudio username and password
  • Setup logs in Amazon CloudWatch Logs (if needed for additional troubleshooting)
  • Amazon EC2 Systems Manager agent, which makes it easy to manage and patch

All AWS resources are created in the US-East-1 Region. To avoid cross-region data transfer fees, launch the CloudFormation stack in the same region. To check the availability of Athena in other regions, see Region Table.

Log in to RStudio

The instance security group has been automatically configured to allow incoming connections on the RStudio port 8787 from any source internet address. You can edit the security group to restrict source IP access. If you have trouble connecting, ensure that port 8787 isn’t blocked by subnet network ACLS or by your outgoing proxy/firewall.

  1. In the CloudFormation stack, choose Outputs, Value, and then open the RStudio URL. You might need to wait for a few minutes until the instance has been launched.
  2. Log in to RStudio with the and password you provided during setup.

Install R packages

Next, install the required R packages from the RStudio console. You can download the R notebook file containing just the code.

#install pacman – a handy package manager for managing installs
if("pacman" %in% rownames(installed.packages()) == FALSE)
h2o.init(nthreads = -1)
##  Connection successful!
## R is connected to the H2O cluster: 
##     H2O cluster uptime:         2 hours 42 minutes 
##     H2O cluster version: 
##     H2O cluster version age:    4 months and 4 days !!! 
##     H2O cluster name:           H2O_started_from_R_rstudio_hjx881 
##     H2O cluster total nodes:    1 
##     H2O cluster total memory:   3.30 GB 
##     H2O cluster total cores:    4 
##     H2O cluster allowed cores:  4 
##     H2O cluster healthy:        TRUE 
##     H2O Connection ip:          localhost 
##     H2O Connection port:        54321 
##     H2O Connection proxy:       NA 
##     H2O Internal Security:      FALSE 
##     R Version:                  R version 3.3.3 (2017-03-06)
## Warning in h2o.clusterInfo(): 
## Your H2O cluster version is too old (4 months and 4 days)!
## Please download and install the latest version from http://h2o.ai/download/
#install aws sdk if not present (pre-requisite for using Athena with an IAM role)
if (!aws_sdk_present()) {


Connect to Athena

Next, establish a connection to Athena from RStudio, using an IAM role associated with your EC2 instance. Use ATHENABUCKET to specify the S3 staging directory.

URL <- 'https://s3.amazonaws.com/athena-downloads/drivers/AthenaJDBC41-1.0.1.jar'
fil <- basename(URL)
#download the file into current working directory
if (!file.exists(fil)) download.file(URL, fil)
#verify that the file has been downloaded successfully
## [1] "AthenaJDBC41-1.0.1.jar"
drv <- JDBC(driverClass="com.amazonaws.athena.jdbc.AthenaDriver", fil, identifier.quote="'")

con <- jdbcConnection <- dbConnect(drv, 'jdbc:awsathena://athena.us-east-1.amazonaws.com:443/',

Verify the connection. The results returned depend on your specific Athena setup.

## <JDBCConnection>
##  [1] "gdelt"               "wikistats"           "elb_logs_raw_native"
##  [4] "twitter"             "twitter2"            "usermovieratings"   
##  [7] "eventcodes"          "events"              "billboard"          
## [10] "billboardtop10"      "elb_logs"            "gdelthist"          
## [13] "gdeltmaster"         "twitter"             "twitter3"

Create a dataset

For this analysis, you use a sample dataset combining information from Billboard and Wikipedia with Echo Nest data in the Million Songs Dataset. Upload this dataset into your own S3 bucket. The table below provides a description of the fields used in this dataset.

Field Description
year Year that song was released
songtitle Title of the song
artistname Name of the song artist
songid Unique identifier for the song
artistid Unique identifier for the song artist
timesignature Variable estimating the time signature of the song
timesignature_confidence Confidence in the estimate for the timesignature
loudness Continuous variable indicating the average amplitude of the audio in decibels
tempo Variable indicating the estimated beats per minute of the song
tempo_confidence Confidence in the estimate for tempo
key Variable with twelve levels indicating the estimated key of the song (C, C#, B)
key_confidence Confidence in the estimate for key
energy Variable that represents the overall acoustic energy of the song, using a mix of features such as loudness
pitch Continuous variable that indicates the pitch of the song
timbre_0_min thru timbre_11_min Variables that indicate the minimum values over all segments for each of the twelve values in the timbre vector
timbre_0_max thru timbre_11_max Variables that indicate the maximum values over all segments for each of the twelve values in the timbre vector
top10 Indicator for whether or not the song made it to the Top 10 of the Billboard charts (1 if it was in the top 10, and 0 if not)

Create an Athena table based on the dataset

In the Athena console, select the default database, sampled, or create a new database.

Run the following create table statement.

create external table if not exists billboard
year int,
songtitle string,
artistname string,
songID string,
artistID string,
timesignature int,
timesignature_confidence double,
loudness double,
tempo double,
tempo_confidence double,
key int,
key_confidence double,
energy double,
pitch double,
timbre_0_min double,
timbre_0_max double,
timbre_1_min double,
timbre_1_max double,
timbre_2_min double,
timbre_2_max double,
timbre_3_min double,
timbre_3_max double,
timbre_4_min double,
timbre_4_max double,
timbre_5_min double,
timbre_5_max double,
timbre_6_min double,
timbre_6_max double,
timbre_7_min double,
timbre_7_max double,
timbre_8_min double,
timbre_8_max double,
timbre_9_min double,
timbre_9_max double,
timbre_10_min double,
timbre_10_max double,
timbre_11_min double,
timbre_11_max double,
Top10 int
LOCATION 's3://aws-bigdata-blog/artifacts/predict-billboard/data'

Inspect the table definition for the ‘billboard’ table that you have created. If you chose a database other than sampledb, replace that value with your choice.

dbGetQuery(con, "show create table sampledb.billboard")
##                                      createtab_stmt
## 1       CREATE EXTERNAL TABLE `sampledb.billboard`(
## 2                                       `year` int,
## 3                               `songtitle` string,
## 4                              `artistname` string,
## 5                                  `songid` string,
## 6                                `artistid` string,
## 7                              `timesignature` int,
## 8                `timesignature_confidence` double,
## 9                                `loudness` double,
## 10                                  `tempo` double,
## 11                       `tempo_confidence` double,
## 12                                       `key` int,
## 13                         `key_confidence` double,
## 14                                 `energy` double,
## 15                                  `pitch` double,
## 16                           `timbre_0_min` double,
## 17                           `timbre_0_max` double,
## 18                           `timbre_1_min` double,
## 19                           `timbre_1_max` double,
## 20                           `timbre_2_min` double,
## 21                           `timbre_2_max` double,
## 22                           `timbre_3_min` double,
## 23                           `timbre_3_max` double,
## 24                           `timbre_4_min` double,
## 25                           `timbre_4_max` double,
## 26                           `timbre_5_min` double,
## 27                           `timbre_5_max` double,
## 28                           `timbre_6_min` double,
## 29                           `timbre_6_max` double,
## 30                           `timbre_7_min` double,
## 31                           `timbre_7_max` double,
## 32                           `timbre_8_min` double,
## 33                           `timbre_8_max` double,
## 34                           `timbre_9_min` double,
## 35                           `timbre_9_max` double,
## 36                          `timbre_10_min` double,
## 37                          `timbre_10_max` double,
## 38                          `timbre_11_min` double,
## 39                          `timbre_11_max` double,
## 40                                     `top10` int)
## 41                             ROW FORMAT DELIMITED 
## 42                         FIELDS TERMINATED BY ',' 
## 43                            STORED AS INPUTFORMAT 
## 44       'org.apache.hadoop.mapred.TextInputFormat' 
## 45                                     OUTPUTFORMAT 
## 46  'org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat'
## 47                                        LOCATION
## 48    's3://aws-bigdata-blog/artifacts/predict-billboard/data'
## 49                                  TBLPROPERTIES (
## 50            'transient_lastDdlTime'='1505484133')

Run a sample query

Next, run a sample query to obtain a list of all songs from Janet Jackson that made it to the Billboard Top 10 charts.

dbGetQuery(con, " SELECT songtitle,artistname,top10   FROM sampledb.billboard WHERE lower(artistname) =     'janet jackson' AND top10 = 1")
##                       songtitle    artistname top10
## 1                       Runaway Janet Jackson     1
## 2               Because Of Love Janet Jackson     1
## 3                         Again Janet Jackson     1
## 4                            If Janet Jackson     1
## 5  Love Will Never Do (Without You) Janet Jackson 1
## 6                     Black Cat Janet Jackson     1
## 7               Come Back To Me Janet Jackson     1
## 8                       Alright Janet Jackson     1
## 9                      Escapade Janet Jackson     1
## 10                Rhythm Nation Janet Jackson     1

Determine how many songs in this dataset are specifically from the year 2010.

dbGetQuery(con, " SELECT count(*)   FROM sampledb.billboard WHERE year = 2010")
##   _col0
## 1   373

The sample dataset provides certain song properties of interest that can be analyzed to gauge the impact to the song’s overall popularity. Look at one such property, timesignature, and determine the value that is the most frequent among songs in the database. Timesignature is a measure of the number of beats and the type of note involved.

Running the query directly may result in an error, as shown in the commented lines below. This error is a result of trying to retrieve a large result set over a JDBC connection, which can cause out-of-memory issues at the client level. To address this, reduce the fetch size and run again.

#t<-dbGetQuery(con, " SELECT timesignature FROM sampledb.billboard")
#Note:  Running the preceding query results in the following error: 
#Error in .jcall(rp, "I", "fetch", stride, block): java.sql.SQLException: The requested #fetchSize is more than the allowed value in Athena. Please reduce the fetchSize and try #again. Refer to the Athena documentation for valid fetchSize values.
# Use the dbSendQuery function, reduce the fetch size, and run again
r <- dbSendQuery(con, " SELECT timesignature     FROM sampledb.billboard")
dftimesignature<- fetch(r, n=-1, block=100)
## [1] TRUE
## dftimesignature
##    0    1    3    4    5    7 
##   10  143  503 6787  112   19
## [1] 7574

From the results, observe that 6787 songs have a timesignature of 4.

Next, determine the song with the highest tempo.

dbGetQuery(con, " SELECT songtitle,artistname,tempo   FROM sampledb.billboard WHERE tempo = (SELECT max(tempo) FROM sampledb.billboard) ")
##                   songtitle      artistname   tempo
## 1 Wanna Be Startin' Somethin' Michael Jackson 244.307

Create the training dataset

Your model needs to be trained such that it can learn and make accurate predictions. Split the data into training and test datasets, and create the training dataset first.  This dataset contains all observations from the year 2009 and earlier. You may face the same JDBC connection issue pointed out earlier, so this query uses a fetch size.

#BillboardTrain <- dbGetQuery(con, "SELECT * FROM sampledb.billboard WHERE year <= 2009")
#Running the preceding query results in the following error:-
#Error in .verify.JDBC.result(r, "Unable to retrieve JDBC result set for ", : Unable to retrieve #JDBC result set for SELECT * FROM sampledb.billboard WHERE year <= 2009 (Internal error)
#Follow the same approach as before to address this issue.

r <- dbSendQuery(con, "SELECT * FROM sampledb.billboard WHERE year <= 2009")
BillboardTrain <- fetch(r, n=-1, block=100)
## [1] TRUE
##   year           songtitle artistname timesignature
## 1 2009 The Awkward Goodbye    Athlete             3
## 2 2009        Rubik's Cube    Athlete             3
##   timesignature_confidence loudness   tempo tempo_confidence
## 1                    0.732   -6.320  89.614   0.652
## 2                    0.906   -9.541 117.742   0.542
## [1] 7201

Create the test dataset

BillboardTest <- dbGetQuery(con, "SELECT * FROM sampledb.billboard where year = 2010")
##   year              songtitle        artistname key
## 1 2010 This Is the House That Doubt Built A Day to Remember  11
## 2 2010        Sticks & Bricks A Day to Remember  10
##   key_confidence    energy pitch timbre_0_min
## 1          0.453 0.9666556 0.024        0.002
## 2          0.469 0.9847095 0.025        0.000
## [1] 373

Convert the training and test datasets into H2O dataframes

train.h2o <- as.h2o(BillboardTrain)
  |                                                                 |   0%
  |=================================================================| 100%
test.h2o <- as.h2o(BillboardTest)
  |                                                                 |   0%
  |=================================================================| 100%

Inspect the column names in your H2O dataframes.

##  [1] "year"                     "songtitle"               
##  [3] "artistname"               "songid"                  
##  [5] "artistid"                 "timesignature"           
##  [7] "timesignature_confidence" "loudness"                
##  [9] "tempo"                    "tempo_confidence"        
## [11] "key"                      "key_confidence"          
## [13] "energy"                   "pitch"                   
## [15] "timbre_0_min"             "timbre_0_max"            
## [17] "timbre_1_min"             "timbre_1_max"            
## [19] "timbre_2_min"             "timbre_2_max"            
## [21] "timbre_3_min"             "timbre_3_max"            
## [23] "timbre_4_min"             "timbre_4_max"            
## [25] "timbre_5_min"             "timbre_5_max"            
## [27] "timbre_6_min"             "timbre_6_max"            
## [29] "timbre_7_min"             "timbre_7_max"            
## [31] "timbre_8_min"             "timbre_8_max"            
## [33] "timbre_9_min"             "timbre_9_max"            
## [35] "timbre_10_min"            "timbre_10_max"           
## [37] "timbre_11_min"            "timbre_11_max"           
## [39] "top10"

Create models

You need to designate the independent and dependent variables prior to applying your modeling algorithms. Because you’re trying to predict the ‘top10’ field, this would be your dependent variable and everything else would be independent.

Create your first model using GLM. Because GLM works best with numeric data, you create your model by dropping non-numeric variables. You only use the variables in the dataset that describe the numerical attributes of the song in the logistic regression model. You won’t use these variables:  “year”, “songtitle”, “artistname”, “songid”, or “artistid”.

y.dep <- 39
x.indep <- c(6:38)
##  [1]  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
## [24] 29 30 31 32 33 34 35 36 37 38

Create Model 1: All numeric variables

Create Model 1 with the training dataset, using GLM as the modeling algorithm and H2O’s built-in h2o.glm function.

modelh1 <- h2o.glm( y = y.dep, x = x.indep, training_frame = train.h2o, family = "binomial")
  |                                                                 |   0%
  |=====                                                            |   8%
  |=================================================================| 100%

Measure the performance of Model 1, using H2O’s built-in performance function.

## H2OBinomialMetrics: glm
## MSE:  0.09924684
## RMSE:  0.3150347
## LogLoss:  0.3220267
## Mean Per-Class Error:  0.2380168
## AUC:  0.8431394
## Gini:  0.6862787
## R^2:  0.254663
## Null Deviance:  326.0801
## Residual Deviance:  240.2319
## AIC:  308.2319
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
##          0   1    Error     Rate
## 0      255  59 0.187898  =59/314
## 1       17  42 0.288136   =17/59
## Totals 272 101 0.203753  =76/373
## Maximum Metrics: Maximum metrics at their respective thresholds
##                         metric threshold    value idx
## 1                       max f1  0.192772 0.525000 100
## 2                       max f2  0.124912 0.650510 155
## 3                 max f0point5  0.416258 0.612903  23
## 4                 max accuracy  0.416258 0.879357  23
## 5                max precision  0.813396 1.000000   0
## 6                   max recall  0.037579 1.000000 282
## 7              max specificity  0.813396 1.000000   0
## 8             max absolute_mcc  0.416258 0.455251  23
## 9   max min_per_class_accuracy  0.161402 0.738854 125
## 10 max mean_per_class_accuracy  0.124912 0.765006 155
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or ` 
## [1] 0.8431394

The AUC metric provides insight into how well the classifier is able to separate the two classes. In this case, the value of 0.8431394 indicates that the classification is good. (A value of 0.5 indicates a worthless test, while a value of 1.0 indicates a perfect test.)

Next, inspect the coefficients of the variables in the dataset.

dfmodelh1 <- as.data.frame(h2o.varimp(modelh1))
##                       names coefficients sign
## 1              timbre_0_max  1.290938663  NEG
## 2                  loudness  1.262941934  POS
## 3                     pitch  0.616995941  NEG
## 4              timbre_1_min  0.422323735  POS
## 5              timbre_6_min  0.349016024  NEG
## 6                    energy  0.348092062  NEG
## 7             timbre_11_min  0.307331997  NEG
## 8              timbre_3_max  0.302225619  NEG
## 9             timbre_11_max  0.243632060  POS
## 10             timbre_4_min  0.224233951  POS
## 11             timbre_4_max  0.204134342  POS
## 12             timbre_5_min  0.199149324  NEG
## 13             timbre_0_min  0.195147119  POS
## 14 timesignature_confidence  0.179973904  POS
## 15         tempo_confidence  0.144242598  POS
## 16            timbre_10_max  0.137644568  POS
## 17             timbre_7_min  0.126995955  NEG
## 18            timbre_10_min  0.123851179  POS
## 19             timbre_7_max  0.100031481  NEG
## 20             timbre_2_min  0.096127636  NEG
## 21           key_confidence  0.083115820  POS
## 22             timbre_6_max  0.073712419  POS
## 23            timesignature  0.067241917  POS
## 24             timbre_8_min  0.061301881  POS
## 25             timbre_8_max  0.060041698  POS
## 26                      key  0.056158445  POS
## 27             timbre_3_min  0.050825116  POS
## 28             timbre_9_max  0.033733561  POS
## 29             timbre_2_max  0.030939072  POS
## 30             timbre_9_min  0.020708113  POS
## 31             timbre_1_max  0.014228818  NEG
## 32                    tempo  0.008199861  POS
## 33             timbre_5_max  0.004837870  POS
## 34                                    NA <NA>

Typically, songs with heavier instrumentation tend to be louder (have higher values in the variable “loudness”) and more energetic (have higher values in the variable “energy”). This knowledge is helpful for interpreting the modeling results.

You can make the following observations from the results:

  • The coefficient estimates for the confidence values associated with the time signature, key, and tempo variables are positive. This suggests that higher confidence leads to a higher predicted probability of a Top 10 hit.
  • The coefficient estimate for loudness is positive, meaning that mainstream listeners prefer louder songs with heavier instrumentation.
  • The coefficient estimate for energy is negative, meaning that mainstream listeners prefer songs that are less energetic, which are those songs with light instrumentation.

These coefficients lead to contradictory conclusions for Model 1. This could be due to multicollinearity issues. Inspect the correlation between the variables “loudness” and “energy” in the training set.

## [1] 0.7399067

This number indicates that these two variables are highly correlated, and Model 1 does indeed suffer from multicollinearity. Typically, you associate a value of -1.0 to -0.5 or 1.0 to 0.5 to indicate strong correlation, and a value of 0.1 to 0.1 to indicate weak correlation. To avoid this correlation issue, omit one of these two variables and re-create the models.

You build two variations of the original model:

  • Model 2, in which you keep “energy” and omit “loudness”
  • Model 3, in which you keep “loudness” and omit “energy”

You compare these two models and choose the model with a better fit for this use case.

Create Model 2: Keep energy and omit loudness

##  [1] "year"                     "songtitle"               
##  [3] "artistname"               "songid"                  
##  [5] "artistid"                 "timesignature"           
##  [7] "timesignature_confidence" "loudness"                
##  [9] "tempo"                    "tempo_confidence"        
## [11] "key"                      "key_confidence"          
## [13] "energy"                   "pitch"                   
## [15] "timbre_0_min"             "timbre_0_max"            
## [17] "timbre_1_min"             "timbre_1_max"            
## [19] "timbre_2_min"             "timbre_2_max"            
## [21] "timbre_3_min"             "timbre_3_max"            
## [23] "timbre_4_min"             "timbre_4_max"            
## [25] "timbre_5_min"             "timbre_5_max"            
## [27] "timbre_6_min"             "timbre_6_max"            
## [29] "timbre_7_min"             "timbre_7_max"            
## [31] "timbre_8_min"             "timbre_8_max"            
## [33] "timbre_9_min"             "timbre_9_max"            
## [35] "timbre_10_min"            "timbre_10_max"           
## [37] "timbre_11_min"            "timbre_11_max"           
## [39] "top10"
y.dep <- 39
x.indep <- c(6:7,9:38)
##  [1]  6  7  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
## [24] 30 31 32 33 34 35 36 37 38
modelh2 <- h2o.glm( y = y.dep, x = x.indep, training_frame = train.h2o, family = "binomial")
  |                                                                 |   0%
  |=======                                                          |  10%
  |=================================================================| 100%

Measure the performance of Model 2.

## H2OBinomialMetrics: glm
## MSE:  0.09922606
## RMSE:  0.3150017
## LogLoss:  0.3228213
## Mean Per-Class Error:  0.2490554
## AUC:  0.8431933
## Gini:  0.6863867
## R^2:  0.2548191
## Null Deviance:  326.0801
## Residual Deviance:  240.8247
## AIC:  306.8247
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
##          0  1    Error     Rate
## 0      280 34 0.108280  =34/314
## 1       23 36 0.389831   =23/59
## Totals 303 70 0.152815  =57/373
## Maximum Metrics: Maximum metrics at their respective thresholds
##                         metric threshold    value idx
## 1                       max f1  0.254391 0.558140  69
## 2                       max f2  0.113031 0.647208 157
## 3                 max f0point5  0.413999 0.596026  22
## 4                 max accuracy  0.446250 0.876676  18
## 5                max precision  0.811739 1.000000   0
## 6                   max recall  0.037682 1.000000 283
## 7              max specificity  0.811739 1.000000   0
## 8             max absolute_mcc  0.254391 0.469060  69
## 9   max min_per_class_accuracy  0.141051 0.716561 131
## 10 max mean_per_class_accuracy  0.113031 0.761821 157
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
dfmodelh2 <- as.data.frame(h2o.varimp(modelh2))
##                       names coefficients sign
## 1                     pitch  0.700331511  NEG
## 2              timbre_1_min  0.510270513  POS
## 3              timbre_0_max  0.402059546  NEG
## 4              timbre_6_min  0.333316236  NEG
## 5             timbre_11_min  0.331647383  NEG
## 6              timbre_3_max  0.252425901  NEG
## 7             timbre_11_max  0.227500308  POS
## 8              timbre_4_max  0.210663865  POS
## 9              timbre_0_min  0.208516163  POS
## 10             timbre_5_min  0.202748055  NEG
## 11             timbre_4_min  0.197246582  POS
## 12            timbre_10_max  0.172729619  POS
## 13         tempo_confidence  0.167523934  POS
## 14 timesignature_confidence  0.167398830  POS
## 15             timbre_7_min  0.142450727  NEG
## 16             timbre_8_max  0.093377516  POS
## 17            timbre_10_min  0.090333426  POS
## 18            timesignature  0.085851625  POS
## 19             timbre_7_max  0.083948442  NEG
## 20           key_confidence  0.079657073  POS
## 21             timbre_6_max  0.076426046  POS
## 22             timbre_2_min  0.071957831  NEG
## 23             timbre_9_max  0.071393189  POS
## 24             timbre_8_min  0.070225578  POS
## 25                      key  0.061394702  POS
## 26             timbre_3_min  0.048384697  POS
## 27             timbre_1_max  0.044721121  NEG
## 28                   energy  0.039698433  POS
## 29             timbre_5_max  0.039469064  POS
## 30             timbre_2_max  0.018461133  POS
## 31                    tempo  0.013279926  POS
## 32             timbre_9_min  0.005282143  NEG
## 33                                    NA <NA>

## [1] 0.8431933

You can make the following observations:

  • The AUC metric is 0.8431933.
  • Inspecting the coefficient of the variable energy, Model 2 suggests that songs with high energy levels tend to be more popular. This is as per expectation.
  • As H2O orders variables by significance, the variable energy is not significant in this model.

You can conclude that Model 2 is not ideal for this use , as energy is not significant.

CreateModel 3: Keep loudness but omit energy

##  [1] "year"                     "songtitle"               
##  [3] "artistname"               "songid"                  
##  [5] "artistid"                 "timesignature"           
##  [7] "timesignature_confidence" "loudness"                
##  [9] "tempo"                    "tempo_confidence"        
## [11] "key"                      "key_confidence"          
## [13] "energy"                   "pitch"                   
## [15] "timbre_0_min"             "timbre_0_max"            
## [17] "timbre_1_min"             "timbre_1_max"            
## [19] "timbre_2_min"             "timbre_2_max"            
## [21] "timbre_3_min"             "timbre_3_max"            
## [23] "timbre_4_min"             "timbre_4_max"            
## [25] "timbre_5_min"             "timbre_5_max"            
## [27] "timbre_6_min"             "timbre_6_max"            
## [29] "timbre_7_min"             "timbre_7_max"            
## [31] "timbre_8_min"             "timbre_8_max"            
## [33] "timbre_9_min"             "timbre_9_max"            
## [35] "timbre_10_min"            "timbre_10_max"           
## [37] "timbre_11_min"            "timbre_11_max"           
## [39] "top10"
y.dep <- 39
x.indep <- c(6:12,14:38)
##  [1]  6  7  8  9 10 11 12 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
## [24] 30 31 32 33 34 35 36 37 38
modelh3 <- h2o.glm( y = y.dep, x = x.indep, training_frame = train.h2o, family = "binomial")
  |                                                                 |   0%
  |========                                                         |  12%
  |=================================================================| 100%
## H2OBinomialMetrics: glm
## MSE:  0.0978859
## RMSE:  0.3128672
## LogLoss:  0.3178367
## Mean Per-Class Error:  0.264925
## AUC:  0.8492389
## Gini:  0.6984778
## R^2:  0.2648836
## Null Deviance:  326.0801
## Residual Deviance:  237.1062
## AIC:  303.1062
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
##          0  1    Error     Rate
## 0      286 28 0.089172  =28/314
## 1       26 33 0.440678   =26/59
## Totals 312 61 0.144772  =54/373
## Maximum Metrics: Maximum metrics at their respective thresholds
##                         metric threshold    value idx
## 1                       max f1  0.273799 0.550000  60
## 2                       max f2  0.125503 0.663265 155
## 3                 max f0point5  0.435479 0.628931  24
## 4                 max accuracy  0.435479 0.882038  24
## 5                max precision  0.821606 1.000000   0
## 6                   max recall  0.038328 1.000000 280
## 7              max specificity  0.821606 1.000000   0
## 8             max absolute_mcc  0.435479 0.471426  24
## 9   max min_per_class_accuracy  0.173693 0.745763 120
## 10 max mean_per_class_accuracy  0.125503 0.775073 155
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
dfmodelh3 <- as.data.frame(h2o.varimp(modelh3))
##                       names coefficients sign
## 1              timbre_0_max 1.216621e+00  NEG
## 2                  loudness 9.780973e-01  POS
## 3                     pitch 7.249788e-01  NEG
## 4              timbre_1_min 3.891197e-01  POS
## 5              timbre_6_min 3.689193e-01  NEG
## 6             timbre_11_min 3.086673e-01  NEG
## 7              timbre_3_max 3.025593e-01  NEG
## 8             timbre_11_max 2.459081e-01  POS
## 9              timbre_4_min 2.379749e-01  POS
## 10             timbre_4_max 2.157627e-01  POS
## 11             timbre_0_min 1.859531e-01  POS
## 12             timbre_5_min 1.846128e-01  NEG
## 13 timesignature_confidence 1.729658e-01  POS
## 14             timbre_7_min 1.431871e-01  NEG
## 15            timbre_10_max 1.366703e-01  POS
## 16            timbre_10_min 1.215954e-01  POS
## 17         tempo_confidence 1.183698e-01  POS
## 18             timbre_2_min 1.019149e-01  NEG
## 19           key_confidence 9.109701e-02  POS
## 20             timbre_7_max 8.987908e-02  NEG
## 21             timbre_6_max 6.935132e-02  POS
## 22             timbre_8_max 6.878241e-02  POS
## 23            timesignature 6.120105e-02  POS
## 24                      key 5.814805e-02  POS
## 25             timbre_8_min 5.759228e-02  POS
## 26             timbre_1_max 2.930285e-02  NEG
## 27             timbre_9_max 2.843755e-02  POS
## 28             timbre_3_min 2.380245e-02  POS
## 29             timbre_2_max 1.917035e-02  POS
## 30             timbre_5_max 1.715813e-02  POS
## 31                    tempo 1.364418e-02  NEG
## 32             timbre_9_min 8.463143e-05  NEG
## 33                                    NA <NA>
## Warning in h2o.find_row_by_threshold(object, t): Could not find exact
## threshold: 0.5 for this set of metrics; using closest threshold found:
## 0.501855569251422. Run `h2o.predict` and apply your desired threshold on a
## probability column.
## [[1]]
## [1] 0.2033898
## [1] 0.8492389

You can make the following observations:

  • The AUC metric is 0.8492389.
  • From the confusion matrix, the model correctly predicts that 33 songs will be top 10 hits (true positives). However, it has 26 false positives (songs that the model predicted would be Top 10 hits, but ended up not being Top 10 hits).
  • Loudness has a positive coefficient estimate, meaning that this model predicts that songs with heavier instrumentation tend to be more popular. This is the same conclusion from Model 2.
  • Loudness is significant in this model.

Overall, Model 3 predicts a higher number of top 10 hits with an accuracy rate that is acceptable. To choose the best fit for production runs, record labels should consider the following factors:

  • Desired model accuracy at a given threshold
  • Number of correct predictions for top10 hits
  • Tolerable number of false positives or false negatives

Next, make predictions using Model 3 on the test dataset.

predict.regh <- h2o.predict(modelh3, test.h2o)
  |                                                                 |   0%
  |=================================================================| 100%
##   predict        p0          p1
## 1       0 0.9654739 0.034526052
## 2       0 0.9654748 0.034525236
## 3       0 0.9635547 0.036445318
## 4       0 0.9343579 0.065642149
## 5       0 0.9978334 0.002166601
## 6       0 0.9779949 0.022005078
## [373 rows x 3 columns]
##   predict
## 1       0
## 2       0
## 3       0
## 4       0
## 5       0
## 6       0
## [373 rows x 1 column]
#Rename the predicted column 
colnames(dpr)[colnames(dpr) == 'predict'] <- 'predict_top10'
##   0   1 
## 312  61

The first set of output results specifies the probabilities associated with each predicted observation.  For example, observation 1 is 96.54739% likely to not be a Top 10 hit, and 3.4526052% likely to be a Top 10 hit (predict=1 indicates Top 10 hit and predict=0 indicates not a Top 10 hit).  The second set of results list the actual predictions made.  From the third set of results, this model predicts that 61 songs will be top 10 hits.

Compute the baseline accuracy, by assuming that the baseline predicts the most frequent outcome, which is that most songs are not Top 10 hits.

##   0   1 
## 314  59

Now observe that the baseline model would get 314 observations correct, and 59 wrong, for an accuracy of 314/(314+59) = 0.8418231.

It seems that Model 3, with an accuracy of 0.8552, provides you with a small improvement over the baseline model. But is this model useful for record labels?

View the two models from an investment perspective:

  • A production company is interested in investing in songs that are more likely to make it to the Top 10. The company’s objective is to minimize the risk of financial losses attributed to investing in songs that end up unpopular.
  • How many songs does Model 3 correctly predict as a Top 10 hit in 2010? Looking at the confusion matrix, you see that it predicts 33 top 10 hits correctly at an optimal threshold, which is more than half the number
  • It will be more useful to the record label if you can provide the production company with a list of songs that are highly likely to end up in the Top 10.
  • The baseline model is not useful, as it simply does not label any song as a hit.

Considering the three models built so far, you can conclude that Model 3 proves to be the best investment choice for the record label.

GBM model

H2O provides you with the ability to explore other learning models, such as GBM and deep learning. Explore building a model using the GBM technique, using the built-in h2o.gbm function.

Before you do this, you need to convert the target variable to a factor for multinomial classification techniques.

gbm.modelh <- h2o.gbm(y=y.dep, x=x.indep, training_frame = train.h2o, ntrees = 500, max_depth = 4, learn_rate = 0.01, seed = 1122,distribution="multinomial")
  |                                                                 |   0%
  |===                                                              |   5%
  |=====                                                            |   7%
  |======                                                           |   9%
  |=======                                                          |  10%
  |======================                                           |  33%
  |=====================================                            |  56%
  |====================================================             |  79%
  |================================================================ |  98%
  |=================================================================| 100%
## H2OBinomialMetrics: gbm
## MSE:  0.09860778
## RMSE:  0.3140188
## LogLoss:  0.3206876
## Mean Per-Class Error:  0.2120263
## AUC:  0.8630573
## Gini:  0.7261146
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
##          0  1    Error     Rate
## 0      266 48 0.152866  =48/314
## 1       16 43 0.271186   =16/59
## Totals 282 91 0.171582  =64/373
## Maximum Metrics: Maximum metrics at their respective thresholds
##                       metric threshold    value idx
## 1                     max f1  0.189757 0.573333  90
## 2                     max f2  0.130895 0.693717 145
## 3               max f0point5  0.327346 0.598802  26
## 4               max accuracy  0.442757 0.876676  14
## 5              max precision  0.802184 1.000000   0
## 6                 max recall  0.049990 1.000000 284
## 7            max specificity  0.802184 1.000000   0
## 8           max absolute_mcc  0.169135 0.496486 104
## 9 max min_per_class_accuracy  0.169135 0.796610 104
## 10 max mean_per_class_accuracy  0.169135 0.805948 104
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `
## Warning in h2o.find_row_by_threshold(object, t): Could not find exact
## threshold: 0.5 for this set of metrics; using closest threshold found:
## 0.501205344484314. Run `h2o.predict` and apply your desired threshold on a
## probability column.
## [[1]]
## [1] 0.1355932
## [1] 0.8630573

This model correctly predicts 43 top 10 hits, which is 10 more than the number predicted by Model 3. Moreover, the AUC metric is higher than the one obtained from Model 3.

As seen above, H2O’s API provides the ability to obtain key statistical measures required to analyze the models easily, using several built-in functions. The record label can experiment with different parameters to arrive at the model that predicts the maximum number of Top 10 hits at the desired level of accuracy and threshold.

H2O also allows you to experiment with deep learning models. Deep learning models have the ability to learn features implicitly, but can be more expensive computationally.

Now, create a deep learning model with the h2o.deeplearning function, using the same training and test datasets created before. The time taken to run this model depends on the type of EC2 instance chosen for this purpose.  For models that require more computation, consider using accelerated computing instances such as the P2 instance type.

  dlearning.modelh <- h2o.deeplearning(y = y.dep,
                                      x = x.indep,
                                      training_frame = train.h2o,
                                      epoch = 250,
                                      hidden = c(250,250),
                                      activation = "Rectifier",
                                      seed = 1122,
  |                                                                 |   0%
  |===                                                              |   4%
  |=====                                                            |   8%
  |========                                                         |  12%
  |==========                                                       |  16%
  |=============                                                    |  20%
  |================                                                 |  24%
  |==================                                               |  28%
  |=====================                                            |  32%
  |=======================                                          |  36%
  |==========================                                       |  40%
  |=============================                                    |  44%
  |===============================                                  |  48%
  |==================================                               |  52%
  |====================================                             |  56%
  |=======================================                          |  60%
  |==========================================                       |  64%
  |============================================                     |  68%
  |===============================================                  |  72%
  |=================================================                |  76%
  |====================================================             |  80%
  |=======================================================          |  84%
  |=========================================================        |  88%
  |============================================================     |  92%
  |==============================================================   |  96%
  |=================================================================| 100%
##    user  system elapsed 
##   1.216   0.020 166.508
## H2OBinomialMetrics: deeplearning
## MSE:  0.1678359
## RMSE:  0.4096778
## LogLoss:  1.86509
## Mean Per-Class Error:  0.3433013
## AUC:  0.7568822
## Gini:  0.5137644
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
##          0  1    Error     Rate
## 0      290 24 0.076433  =24/314
## 1       36 23 0.610169   =36/59
## Totals 326 47 0.160858  =60/373
## Maximum Metrics: Maximum metrics at their respective thresholds
##                       metric threshold    value idx
## 1                     max f1  0.826267 0.433962  46
## 2                     max f2  0.000000 0.588235 239
## 3               max f0point5  0.999929 0.511811  16
## 4               max accuracy  0.999999 0.865952  10
## 5              max precision  1.000000 1.000000   0
## 6                 max recall  0.000000 1.000000 326
## 7            max specificity  1.000000 1.000000   0
## 8           max absolute_mcc  0.999929 0.363219  16
## 9 max min_per_class_accuracy  0.000004 0.662420 145
## 10 max mean_per_class_accuracy  0.000000 0.685334 224
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## Warning in h2o.find_row_by_threshold(object, t): Could not find exact
## threshold: 0.5 for this set of metrics; using closest threshold found:
## 0.496293348880151. Run `h2o.predict` and apply your desired threshold on a
## probability column.
## [[1]]
## [1] 0.3898305
## [1] 0.7568822

The AUC metric for this model is 0.7568822, which is less than what you got from the earlier models. I recommend further experimentation using different hyper parameters, such as the learning rate, epoch or the number of hidden layers.

H2O’s built-in functions provide many key statistical measures that can help measure model performance. Here are some of these key terms.

Metric Description
Sensitivity Measures the proportion of positives that have been correctly identified. It is also called the true positive rate, or recall.
Specificity Measures the proportion of negatives that have been correctly identified. It is also called the true negative rate.
Threshold Cutoff point that maximizes specificity and sensitivity. While the model may not provide the highest prediction at this point, it would not be biased towards positives or negatives.
Precision The fraction of the documents retrieved that are relevant to the information needed, for example, how many of the positively classified are relevant

Provides insight into how well the classifier is able to separate the two classes. The implicit goal is to deal with situations where the sample distribution is highly skewed, with a tendency to overfit to a single class.

0.90 – 1 = excellent (A)

0.8 – 0.9 = good (B)

0.7 – 0.8 = fair (C)

.6 – 0.7 = poor (D)

0.5 – 0.5 = fail (F)

Here’s a summary of the metrics generated from H2O’s built-in functions for the three models that produced useful results.

Metric Model 3 GBM Model Deep Learning Model



















1.0 1.0





1.0 1.0





0.2033898 0.1355932



AUC 0.8492389 0.8630573 0.756882

Note: ‘t’ denotes threshold.

Your options at this point could be narrowed down to Model 3 and the GBM model, based on the AUC and accuracy metrics observed earlier.  If the slightly lower accuracy of the GBM model is deemed acceptable, the record label can choose to go to production with the GBM model, as it can predict a higher number of Top 10 hits.  The AUC metric for the GBM model is also higher than that of Model 3.

Record labels can experiment with different learning techniques and parameters before arriving at a model that proves to be the best fit for their business. Because deep learning models can be computationally expensive, record labels can choose more powerful EC2 instances on AWS to run their experiments faster.


In this post, I showed how the popular music industry can use analytics to predict the type of songs that make the Top 10 Billboard charts. By running H2O’s scalable machine learning platform on AWS, data scientists can easily experiment with multiple modeling techniques and interactively query the data using Amazon Athena, without having to manage the underlying infrastructure. This helps record labels make critical decisions on the type of artists and songs to promote in a timely fashion, thereby increasing sales and revenue.

If you have questions or suggestions, please comment below.

Additional Reading

Learn how to build and explore a simple geospita simple GEOINT application using SparkR.

About the Authors

gopalGopal Wunnava is a Partner Solution Architect with the AWS GSI Team. He works with partners and customers on big data engagements, and is passionate about building analytical solutions that drive business capabilities and decision making. In his spare time, he loves all things sports and movies related and is fond of old classics like Asterix, Obelix comics and Hitchcock movies.



Bob Strahan, a Senior Consultant with AWS Professional Services, contributed to this post.



Clean up Your Container Images with Amazon ECR Lifecycle Policies

Post Syndicated from Nathan Taber original https://aws.amazon.com/blogs/compute/clean-up-your-container-images-with-amazon-ecr-lifecycle-policies/

This post comes from the desk of Brent Langston.

Starting today, customers can keep their container image repositories tidy by automatically removing old or unused images using lifecycle policies, now available as part of Amazon E2 Container Repository (Amazon ECR).

Amazon ECR is a fully managed Docker container registry that makes it easy to store manage and deploy Docker container images without worrying about the typical challenges of scaling a service to handle pulling hundreds of images at one time. This scale means that development teams using Amazon ECR actively often find that their repositories fill up with many container image versions. This makes it difficult to find the code changes that matter and incurs unnecessary storage costs. Previously, cleaning up your repository meant spending time to manually delete old images, or writing and executing scripts.

Now, lifecycle policies allow you to define a set of rules to remove old container images automatically. You can also preview rules to see exactly which container images are affected when the rule runs. This allows repositories to be better organized, makes it easier to find the code revisions that matter, and lowers storage costs.

Look at how lifecycle policies work.

Ground Rules

One of the biggest benefits of deploying code in containers is the ability to quickly and easily roll back to a previous version. You can deploy with less risk because, if something goes wrong, it is easy to revert back to the previous container version and know that your application will run like it did before the failed deployment. Most people probably never roll back past a few versions. If your situation is similar, then one simple lifecycle rule might be to just keep the last 30 images.

Last 30 Images

In your ECR registry, choose Dry-Run Lifecycle Rules, Add.

  • For Image Status, select Untagged.
  • Under Match criteria, for Count Type, enter Image Count More Than.
  • For Count Number, enter 30.
  • For Rule action, choose expire.

Choose Save. To see which images would be cleaned up, Save and dry-run rules.

Of course, there are teams who, for compliance reasons, might prefer to keep certain images for a period of time, rather than keeping by count. For that situation, you can choose to clean up images older than 90 days.

Last 90 Days

Select the rule that you just created and choose Edit. Change the parameters to keep only 90 days of untagged images:

  • Under Match criteria, for Count Type, enter Since Image Pushed
  • For Count Number, enter 90.
  • For Count Unit, enter days.


Certainly 90 days is an arbitrary timeframe, and your team might have policies in place that would require a longer timeframe for certain kinds of images. If that’s the case, but you still want to continue with the spring cleaning, you can consider getting rid of images that are tag prefixed.

Here is the list of rules I came up with to groom untagged, development, staging, and production images:

  • Remove untagged images over 90 days old
  • Remove development tagged images over 90 days old
  • Remove staging tagged images over 180 days old
  • Remove production tagged images over 1 year old

As you can see, the new Amazon ECR lifecycle policies are powerful, and help you easily keep the images you need, while cleaning out images you may never use again. This feature is available starting today, in all regions where Amazon ECR is available, at no extra charge. For more information, see Amazon ECR Lifecycle Policies in the AWS technical documentation.

— Brent

"Responsible encryption" fallacies

Post Syndicated from Robert Graham original http://blog.erratasec.com/2017/10/responsible-encryption-fallacies.html

Deputy Attorney General Rod Rosenstein gave a speech recently calling for “Responsible Encryption” (aka. “Crypto Backdoors”). It’s full of dangerous ideas that need to be debunked.

The importance of law enforcement

The first third of the speech talks about the importance of law enforcement, as if it’s the only thing standing between us and chaos. It cites the 2016 Mirai attacks as an example of the chaos that will only get worse without stricter law enforcement.

But the Mira case demonstrated the opposite, how law enforcement is not needed. They made no arrests in the case. A year later, they still haven’t a clue who did it.

Conversely, we technologists have fixed the major infrastructure issues. Specifically, those affected by the DNS outage have moved to multiple DNS providers, including a high-capacity DNS provider like Google and Amazon who can handle such large attacks easily.

In other words, we the people fixed the major Mirai problem, and law-enforcement didn’t.

Moreover, instead being a solution to cyber threats, law enforcement has become a threat itself. The DNC didn’t have the FBI investigate the attacks from Russia likely because they didn’t want the FBI reading all their files, finding wrongdoing by the DNC. It’s not that they did anything actually wrong, but it’s more like that famous quote from Richelieu “Give me six words written by the most honest of men and I’ll find something to hang him by”. Give all your internal emails over to the FBI and I’m certain they’ll find something to hang you by, if they want.
Or consider the case of Andrew Auernheimer. He found AT&T’s website made public user accounts of the first iPad, so he copied some down and posted them to a news site. AT&T had denied the problem, so making the problem public was the only way to force them to fix it. Such access to the website was legal, because AT&T had made the data public. However, prosecutors disagreed. In order to protect the powerful, they twisted and perverted the law to put Auernheimer in jail.

It’s not that law enforcement is bad, it’s that it’s not the unalloyed good Rosenstein imagines. When law enforcement becomes the thing Rosenstein describes, it means we live in a police state.

Where law enforcement can’t go

Rosenstein repeats the frequent claim in the encryption debate:

Our society has never had a system where evidence of criminal wrongdoing was totally impervious to detection

Of course our society has places “impervious to detection”, protected by both legal and natural barriers.

An example of a legal barrier is how spouses can’t be forced to testify against each other. This barrier is impervious.

A better example, though, is how so much of government, intelligence, the military, and law enforcement itself is impervious. If prosecutors could gather evidence everywhere, then why isn’t Rosenstein prosecuting those guilty of CIA torture?

Oh, you say, government is a special exception. If that were the case, then why did Rosenstein dedicate a precious third of his speech discussing the “rule of law” and how it applies to everyone, “protecting people from abuse by the government”. It obviously doesn’t, there’s one rule of government and a different rule for the people, and the rule for government means there’s lots of places law enforcement can’t go to gather evidence.

Likewise, the crypto backdoor Rosenstein is demanding for citizens doesn’t apply to the President, Congress, the NSA, the Army, or Rosenstein himself.

Then there are the natural barriers. The police can’t read your mind. They can only get the evidence that is there, like partial fingerprints, which are far less reliable than full fingerprints. They can’t go backwards in time.

I mention this because encryption is a natural barrier. It’s their job to overcome this barrier if they can, to crack crypto and so forth. It’s not our job to do it for them.

It’s like the camera that increasingly comes with TVs for video conferencing, or the microphone on Alexa-style devices that are always recording. This suddenly creates evidence that the police want our help in gathering, such as having the camera turned on all the time, recording to disk, in case the police later gets a warrant, to peer backward in time what happened in our living rooms. The “nothing is impervious” argument applies here as well. And it’s equally bogus here. By not helping police by not recording our activities, we aren’t somehow breaking some long standing tradit

And this is the scary part. It’s not that we are breaking some ancient tradition that there’s no place the police can’t go (with a warrant). Instead, crypto backdoors breaking the tradition that never before have I been forced to help them eavesdrop on me, even before I’m a suspect, even before any crime has been committed. Sure, laws like CALEA force the phone companies to help the police against wrongdoers — but here Rosenstein is insisting I help the police against myself.

Balance between privacy and public safety

Rosenstein repeats the frequent claim that encryption upsets the balance between privacy/safety:

Warrant-proof encryption defeats the constitutional balance by elevating privacy above public safety.

This is laughable, because technology has swung the balance alarmingly in favor of law enforcement. Far from “Going Dark” as his side claims, the problem we are confronted with is “Going Light”, where the police state monitors our every action.

You are surrounded by recording devices. If you walk down the street in town, outdoor surveillance cameras feed police facial recognition systems. If you drive, automated license plate readers can track your route. If you make a phone call or use a credit card, the police get a record of the transaction. If you stay in a hotel, they demand your ID, for law enforcement purposes.

And that’s their stuff, which is nothing compared to your stuff. You are never far from a recording device you own, such as your mobile phone, TV, Alexa/Siri/OkGoogle device, laptop. Modern cars from the last few years increasingly have always-on cell connections and data recorders that record your every action (and location).

Even if you hike out into the country, when you get back, the FBI can subpoena your GPS device to track down your hidden weapon’s cache, or grab the photos from your camera.

And this is all offline. So much of what we do is now online. Of the photographs you own, fewer than 1% are printed out, the rest are on your computer or backed up to the cloud.

Your phone is also a GPS recorder of your exact position all the time, which if the government wins the Carpenter case, they police can grab without a warrant. Tagging all citizens with a recording device of their position is not “balance” but the premise for a novel more dystopic than 1984.

If suspected of a crime, which would you rather the police searched? Your person, houses, papers, and physical effects? Or your mobile phone, computer, email, and online/cloud accounts?

The balance of privacy and safety has swung so far in favor of law enforcement that rather than debating whether they should have crypto backdoors, we should be debating how to add more privacy protections.

“But it’s not conclusive”

Rosenstein defends the “going light” (“Golden Age of Surveillance”) by pointing out it’s not always enough for conviction. Nothing gives a conviction better than a person’s own words admitting to the crime that were captured by surveillance. This other data, while copious, often fails to convince a jury beyond a reasonable doubt.
This is nonsense. Police got along well enough before the digital age, before such widespread messaging. They solved terrorist and child abduction cases just fine in the 1980s. Sure, somebody’s GPS location isn’t by itself enough — until you go there and find all the buried bodies, which leads to a conviction. “Going dark” imagines that somehow, the evidence they’ve been gathering for centuries is going away. It isn’t. It’s still here, and matches up with even more digital evidence.
Conversely, a person’s own words are not as conclusive as you think. There’s always missing context. We quickly get back to the Richelieu “six words” problem, where captured communications are twisted to convict people, with defense lawyers trying to untwist them.

Rosenstein’s claim may be true, that a lot of criminals will go free because the other electronic data isn’t convincing enough. But I’d need to see that claim backed up with hard studies, not thrown out for emotional impact.

Terrorists and child molesters

You can always tell the lack of seriousness of law enforcement when they bring up terrorists and child molesters.
To be fair, sometimes we do need to talk about terrorists. There are things unique to terrorism where me may need to give government explicit powers to address those unique concerns. For example, the NSA buys mobile phone 0day exploits in order to hack terrorist leaders in tribal areas. This is a good thing.
But when terrorists use encryption the same way everyone else does, then it’s not a unique reason to sacrifice our freedoms to give the police extra powers. Either it’s a good idea for all crimes or no crimes — there’s nothing particular about terrorism that makes it an exceptional crime. Dead people are dead. Any rational view of the problem relegates terrorism to be a minor problem. More citizens have died since September 8, 2001 from their own furniture than from terrorism. According to studies, the hot water from the tap is more of a threat to you than terrorists.
Yes, government should do what they can to protect us from terrorists, but no, it’s not so bad of a threat that requires the imposition of a military/police state. When people use terrorism to justify their actions, it’s because they trying to form a military/police state.
A similar argument works with child porn. Here’s the thing: the pervs aren’t exchanging child porn using the services Rosenstein wants to backdoor, like Apple’s Facetime or Facebook’s WhatsApp. Instead, they are exchanging child porn using custom services they build themselves.
Again, I’m (mostly) on the side of the FBI. I support their idea of buying 0day exploits in order to hack the web browsers of visitors to the secret “PlayPen” site. This is something that’s narrow to this problem and doesn’t endanger the innocent. On the other hand, their calls for crypto backdoors endangers the innocent while doing effectively nothing to address child porn.
Terrorists and child molesters are a clichéd, non-serious excuse to appeal to our emotions to give up our rights. We should not give in to such emotions.

Definition of “backdoor”

Rosenstein claims that we shouldn’t call backdoors “backdoors”:

No one calls any of those functions [like key recovery] a “back door.”  In fact, those capabilities are marketed and sought out by many users.

He’s partly right in that we rarely refer to PGP’s key escrow feature as a “backdoor”.

But that’s because the term “backdoor” refers less to how it’s done and more to who is doing it. If I set up a recovery password with Apple, I’m the one doing it to myself, so we don’t call it a backdoor. If it’s the police, spies, hackers, or criminals, then we call it a “backdoor” — even it’s identical technology.

Wikipedia uses the key escrow feature of the 1990s Clipper Chip as a prime example of what everyone means by “backdoor“. By “no one”, Rosenstein is including Wikipedia, which is obviously incorrect.

Though in truth, it’s not going to be the same technology. The needs of law enforcement are different than my personal key escrow/backup needs. In particular, there are unsolvable problems, such as a backdoor that works for the “legitimate” law enforcement in the United States but not for the “illegitimate” police states like Russia and China.

I feel for Rosenstein, because the term “backdoor” does have a pejorative connotation, which can be considered unfair. But that’s like saying the word “murder” is a pejorative term for killing people, or “torture” is a pejorative term for torture. The bad connotation exists because we don’t like government surveillance. I mean, honestly calling this feature “government surveillance feature” is likewise pejorative, and likewise exactly what it is that we are talking about.


Rosenstein focuses his arguments on “providers”, like Snapchat or Apple. But this isn’t the question.

The question is whether a “provider” like Telegram, a Russian company beyond US law, provides this feature. Or, by extension, whether individuals should be free to install whatever software they want, regardless of provider.

Telegram is a Russian company that provides end-to-end encryption. Anybody can download their software in order to communicate so that American law enforcement can’t eavesdrop. They aren’t going to put in a backdoor for the U.S. If we succeed in putting backdoors in Apple and WhatsApp, all this means is that criminals are going to install Telegram.

If the, for some reason, the US is able to convince all such providers (including Telegram) to install a backdoor, then it still doesn’t solve the problem, as uses can just build their own end-to-end encryption app that has no provider. It’s like email: some use the major providers like GMail, others setup their own email server.

Ultimately, this means that any law mandating “crypto backdoors” is going to target users not providers. Rosenstein tries to make a comparison with what plain-old telephone companies have to do under old laws like CALEA, but that’s not what’s happening here. Instead, for such rules to have any effect, they have to punish users for what they install, not providers.

This continues the argument I made above. Government backdoors is not something that forces Internet services to eavesdrop on us — it forces us to help the government spy on ourselves.
Rosenstein tries to address this by pointing out that it’s still a win if major providers like Apple and Facetime are forced to add backdoors, because they are the most popular, and some terrorists/criminals won’t move to alternate platforms. This is false. People with good intentions, who are unfairly targeted by a police state, the ones where police abuse is rampant, are the ones who use the backdoored products. Those with bad intentions, who know they are guilty, will move to the safe products. Indeed, Telegram is already popular among terrorists because they believe American services are already all backdoored. 
Rosenstein is essentially demanding the innocent get backdoored while the guilty don’t. This seems backwards. This is backwards.

Apple is morally weak

The reason I’m writing this post is because Rosenstein makes a few claims that cannot be ignored. One of them is how he describes Apple’s response to government insistence on weakening encryption doing the opposite, strengthening encryption. He reasons this happens because:

Of course they [Apple] do. They are in the business of selling products and making money. 

We [the DoJ] use a different measure of success. We are in the business of preventing crime and saving lives. 

He swells in importance. His condescending tone ennobles himself while debasing others. But this isn’t how things work. He’s not some white knight above the peasantry, protecting us. He’s a beat cop, a civil servant, who serves us.

A better phrasing would have been:

They are in the business of giving customers what they want.

We are in the business of giving voters what they want.

Both sides are doing the same, giving people what they want. Yes, voters want safety, but they also want privacy. Rosenstein imagines that he’s free to ignore our demands for privacy as long has he’s fulfilling his duty to protect us. He has explicitly rejected what people want, “we use a different measure of success”. He imagines it’s his job to tell us where the balance between privacy and safety lies. That’s not his job, that’s our job. We, the people (and our representatives), make that decision, and it’s his job is to do what he’s told. His measure of success is how well he fulfills our wishes, not how well he satisfies his imagined criteria.

That’s why those of us on this side of the debate doubt the good intentions of those like Rosenstein. He criticizes Apple for wanting to protect our rights/freedoms, and declare they measure success differently.

They are willing to be vile

Rosenstein makes this argument:

Companies are willing to make accommodations when required by the government. Recent media reports suggest that a major American technology company developed a tool to suppress online posts in certain geographic areas in order to embrace a foreign government’s censorship policies. 

Let me translate this for you:

Companies are willing to acquiesce to vile requests made by police-states. Therefore, they should acquiesce to our vile police-state requests.

It’s Rosenstein who is admitting here is that his requests are those of a police-state.

Constitutional Rights

Rosenstein says:

There is no constitutional right to sell warrant-proof encryption.

Maybe. It’s something the courts will have to decide. There are many 1st, 2nd, 3rd, 4th, and 5th Amendment issues here.
The reason we have the Bill of Rights is because of the abuses of the British Government. For example, they quartered troops in our homes, as a way of punishing us, and as a way of forcing us to help in our own oppression. The troops weren’t there to defend us against the French, but to defend us against ourselves, to shoot us if we got out of line.

And that’s what crypto backdoors do. We are forced to be agents of our own oppression. The principles enumerated by Rosenstein apply to a wide range of even additional surveillance. With little change to his speech, it can equally argue why the constant TV video surveillance from 1984 should be made law.

Let’s go back and look at Apple. It is not some base company exploiting consumers for profit. Apple doesn’t have guns, they cannot make people buy their product. If Apple doesn’t provide customers what they want, then customers vote with their feet, and go buy an Android phone. Apple isn’t providing encryption/security in order to make a profit — it’s giving customers what they want in order to stay in business.
Conversely, if we citizens don’t like what the government does, tough luck, they’ve got the guns to enforce their edicts. We can’t easily vote with our feet and walk to another country. A “democracy” is far less democratic than capitalism. Apple is a minority, selling phones to 45% of the population, and that’s fine, the minority get the phones they want. In a Democracy, where citizens vote on the issue, those 45% are screwed, as the 55% impose their will unwanted onto the remainder.

That’s why we have the Bill of Rights, to protect the 49% against abuse by the 51%. Regardless whether the Supreme Court agrees the current Constitution, it is the sort right that might exist regardless of what the Constitution says. 

Obliged to speak the truth

Here is the another part of his speech that I feel cannot be ignored. We have to discuss this:

Those of us who swear to protect the rule of law have a different motivation.  We are obliged to speak the truth.

The truth is that “going dark” threatens to disable law enforcement and enable criminals and terrorists to operate with impunity.

This is not true. Sure, he’s obliged to say the absolute truth, in court. He’s also obliged to be truthful in general about facts in his personal life, such as not lying on his tax return (the sort of thing that can get lawyers disbarred).

But he’s not obliged to tell his spouse his honest opinion whether that new outfit makes them look fat. Likewise, Rosenstein knows his opinion on public policy doesn’t fall into this category. He can say with impunity that either global warming doesn’t exist, or that it’ll cause a biblical deluge within 5 years. Both are factually untrue, but it’s not going to get him fired.

And this particular claim is also exaggerated bunk. While everyone agrees encryption makes law enforcement’s job harder than with backdoors, nobody honestly believes it can “disable” law enforcement. While everyone agrees that encryption helps terrorists, nobody believes it can enable them to act with “impunity”.

I feel bad here. It’s a terrible thing to question your opponent’s character this way. But Rosenstein made this unavoidable when he clearly, with no ambiguity, put his integrity as Deputy Attorney General on the line behind the statement that “going dark threatens to disable law enforcement and enable criminals and terrorists to operate with impunity”. I feel it’s a bald face lie, but you don’t need to take my word for it. Read his own words yourself and judge his integrity.


Rosenstein’s speech includes repeated references to ideas like “oath”, “honor”, and “duty”. It reminds me of Col. Jessup’s speech in the movie “A Few Good Men”.

If you’ll recall, it was rousing speech, “you want me on that wall” and “you use words like honor as a punchline”. Of course, since he was violating his oath and sending two privates to death row in order to avoid being held accountable, it was Jessup himself who was crapping on the concepts of “honor”, “oath”, and “duty”.

And so is Rosenstein. He imagines himself on that wall, doing albeit terrible things, justified by his duty to protect citizens. He imagines that it’s he who is honorable, while the rest of us not, even has he utters bald faced lies to further his own power and authority.

We activists oppose crypto backdoors not because we lack honor, or because we are criminals, or because we support terrorists and child molesters. It’s because we value privacy and government officials who get corrupted by power. It’s not that we fear Trump becoming a dictator, it’s that we fear bureaucrats at Rosenstein’s level becoming drunk on authority — which Rosenstein demonstrably has. His speech is a long train of corrupt ideas pursuing the same object of despotism — a despotism we oppose.

In other words, we oppose crypto backdoors because it’s not a tool of law enforcement, but a tool of despotism.

Тръмп, лицензиите на NBC, Първата поправка

Post Syndicated from nellyo original https://nellyo.wordpress.com/2017/10/11/nbc/


Президентът Тръмп открито днес поставя въпроса за отнемане на лицензиите на NBC и други критично настроени медии, недоволен от новинарските им емисии.  Отдавна се знае, че Тръмп сочи CNN като производител на фалшиви новини, сега към CNN се добавят и други медии.

Отделен въпрос е кой и как може да отнеме лицензии – това е регулаторът FCC – и то при определени основания – и то не на цели мрежи. Но това не прави заплахата на президента по-малко опасна. Става дума за конституционна разпоредба  – зачитане на свободата на изразяване според Първата поправка на Конституцията на САЩ. Ценност, която и президентите не си позволяват да атакуват.

Filed under: Media Law, US Law

След iOS 11 mobile-only е все по-възможно

Post Syndicated from Йовко Ламбрев original https://yovko.net/ios11/

В края на септември Apple пусна на вода новата версия на мобилната си операционна платформа. И едва ли щях да пиша нарочен пост за това, ако най-значимият белег на iOS 11 някак не остана подценен, вероятно защото е свързан с философията на платформата по отношение на посоката на развитието ѝ, а не с поредните технологични характеристики. А iOS 11 е крайъгълен камък не защото впечатлява с кой знае каква нова визия или подход, а защото дава заявка за пълноценна, самостоятелна операционна система и изглажда пътя към mobile-only работата. Като блести най-вече на iPad – даже не просто блести, а започва да ти се струва, че направо все едно iPad се е преродил отново.

Признавам, че темата ме вълнува, защото си мечтая един ден (и се очертава да е скоро) да не си купувам повече лаптоп, а таблетът да е всичко, което ми е нужно за да върша работата си пълноценно и удобно. Все още не мога да си го позволя, защото има няколко неща, които не мога да свърша с iPad, но те остават все по-малко и по-малко.

Експериментирам да работя само с iPad от години насам, но нищо не ми е давало такава увереност, че един ден това ще е възможно, както промените, които донесе iOS 11.

Всъщност най-голямата благина, която ми дава работата с iPad е… концентрация. Което от своя страна ми носи по-голяма ефективност и съответно повече удовлетворение. Личи от няколко версии насам, като очевидно е мислено отдавна, че многозадачността в iOS е планирана да е далеч по-грижовна към концентрацията в основната задача, с която се предполага да съм зает в момента. Всички други мобилни и десктоп платформи сякаш изпитват перверзно удоволствие да разфокусират вниманието ми с всевъзможни нотификации, чието озаптяване до приемлива норма изисква екстра усилия, които трябва да бъдат положени, за да може човек да свърши нещо. Затова, особено когато пиша или чета внимателно някакъв текст или код, концентрацията ми е ключова, и често в такива моменти предпочитам iPad-а си пред компютъра.

С появата на iOS 11 многозадачността е под още по-голям контрол – като отново най-невъзмутимо мога да продължа да си бъда фокусиран в най-важното, което правя (еднозадачният режим винаги ми е най-любим), но имам и гъвкавост, с която мога да си поделя екрана с други задачи или да оставя комбинации от различни приложения върху един екран „залепени“ и на background с не повече от две докосвания. А това е голямо облекчение в ежедневието с таблет. Това заедно с появата на Dock и усъвършенстваните Split View и Slide Over функционалности ми дава не просто почти пълноценно десктоп усещане, ами изцяло ново такова, което намирам за много по-удобно и ергономично. За което помага и едно приложение, което от скоро е собственост на Apple, но иначе не беше тяхно, а именно Workflow, но за него някой друг път. Сега само ще кажа, че веднъж като го вкусиш и повече не можеш без него.

Другият голям бонус (още от iPad 1, всъщност) е мобилността и факта, че с едно зареждане на батерията мога с часове да работя напълно автономно и безгрижно. Тук с уговорката, че при дълга работа с iPad, особено на бюро, предпочитам да пиша с реална клавиатура – ползвам класическата Apple Magic keyboard.

Някои от тези неща с iPad Pro и наличието на pensil, който пък отключва и други функции, са още по-секси, но понеже нещата на Apple не само работят добре, ами работят и дълго с години, и могат да носят доста време всички обновявания на платформата, текущият ми iPad e още твърде пълноценен за да го сменям с Pro. Но ще държа темата отворена, защото mobile-only подхода ще продължи да занимава вниманието ми и занапред и имам какво да разкажа за няколко различни направления.

И понеже като напиша нещо за Apple, обикновено следва хейт и легенди как с едни други платформи било по-гот – приключвам този текст с едно от любимите ми шеговити клипчета на Apple по въпроса 😉

14 години

Post Syndicated from Йовко Ламбрев original https://yovko.net/14years/

Преди 6 години беше последният път, когато отбелязах рожденния ден на този блог. Оттогава понамалих честотата на писането тук (а и по принцип писането). Сега, още 6 години след това няма да обещавам, че отново ще го зачестя, не само защото няма и сам да си повярвам, но и защото си мисля, че новото темпо е по-близо до сегашното ми аз – говори само, когато има какво да каже (или да изкрещи, което все по-често му се налага). А и ми предстои да пиша скоро на едно друго място и не зная колко теми и сили ще остават за този блог. Ще видим…

Иначе 14 години са ужасно много време. Точно толкова ме нямаше и в Пловдив и междувременно разбрах, че това което наричаме корени, всъщност не съществува, но усилията да се почувстваш отново на мястото си са съвсем реални. Поне за мен.

Вероятно съм останал един от малкото, които още ползваме RSS-четец и с всеки ден, ровейки се в емисиите му, все повече ми липсва не откриването на нови интересни блогъри, защото такива продължават да се появяват, а отсъствието на старите. И индиректната комуникация с тях – с тези които имахме какво да си кажем и да се прочетем – без оковите на клишетата в главите ни, а с широкоскроеността на съзнанията ни. Не с реплики из социалките, а с лични гледни точки, развити в блоговете ни, с повече от няколко бързи изречения.

Подобно на философията slow food може би има потребност и от бавна публицистика – такава лична от първо лице. Поне аз си признавам, че имам потребност от нея.

Иначе няма да спра да пиша тук, колкото и да е рядко… Но да кажа, че липсвате – тези които моят Inoreader листва като inactive feed. И да – усилията, да се почувстваш отново на мястото след дълга пауза са съвсем реални, но си струват.

ЕСПЧ: Защита на тайната на източниците

Post Syndicated from nellyo original https://nellyo.wordpress.com/2017/10/07/echr_sources/

На 5 октомври 2017 е оповестено решение на ЕСПЧ по делото Becker v. Norway (application no. 21272/12)

На Сесилия Бекер, журналистка, е наредено да даде показания за източниците си на информация  в наказателно дело за манипулиране на пазара. Тя е автор на статия относно затрудненото положение на  Норвежката петролна компания и има контакти с определени източници, които отказва да разкрие. Източниците са станали известни по друг начин в хода на разследването. 

Европейският съд по правата на човека   единодушно приема, че има нарушение на член 10 (свобода на изразяване) на Европейската конвенция за правата на човека. Съдът констатира, че журналистическите методи на г-жа Бекер никога не са били поставяни под съмнение и тя не е била обвинена в никаква незаконна дейност.Отказът й да разкрие източника си не е препятствал производството. Съдът не намира достатъчно причини да се настоява г-жа Бекер да свидетелства в конкретните обстоятелства.

Правото   на журналист  на защита на тайната на източниците не отпада автоматично, ако самоличността  на източника по някакви причини е станала известна.

Filed under: Media Law Tagged: еспч

Things Go Better With Step Functions

Post Syndicated from Jeff Barr original https://aws.amazon.com/blogs/aws/things-go-better-with-step-functions/

I often give presentations on Amazon’s culture of innovation, and start out with a slide that features a revealing quote from Amazon founder Jeff Bezos:

I love to sit down with our customers and to learn how we have empowered their creativity and to pursue their dreams. Earlier this year I chatted with Patrick from The Coca-Cola Company in order to learn how they used AWS Step Functions and other AWS services to support the Coke.com Vending Pass program. This program includes drink rewards earned by purchasing products at vending machines equipped to support mobile payments using the Coca-Cola Vending Pass. Participants swipe their NFC-enabled phones to complete an Apple Pay or Android Pay purchase, identifying themselves to the vending machine and earning credit towards future free vending purchases in the process

After the swipe, a combination of SNS topics and AWS Lambda functions initiated a pair of calls to some existing backend code to count the vending points and update the participant’s record. Unfortunately, the backend code was slow to react and had some timing dependencies, leading to missing updates that had the potential to confuse Vending Pass participants. The initial solution to this issue was very simple: modify the Lambda code to include a 90 second delay between the two calls. This solved the problem, but ate up process time for no good reason (billing for the use of Lambda functions is based on the duration of the request, in 100 ms intervals).

In order to make their solution more cost-effective, the team turned to AWS Step Functions, building a very simple state machine. As I wrote in an earlier blog post, Step Functions coordinate the components of distributed applications and microservices at scale, using visual workflows that are easy to build.

Coke built a very simple state machine to simplify their business logic and reduce their costs. Yours can be equally simple, or they can make use of other Step Function features such as sequential and parallel execution and the ability to make decisions and choose alternate states. The Coke state machine looks like this:

The FirstState and the SecondState states (Task states) call the appropriate Lambda functions while Step Functions implements the 90 second delay (a Wait state). This modification simplified their logic and reduced their costs. Here’s how it all fits together:


What’s Next
This initial success led them to take a closer look at serverless computing and to consider using it for other projects. Patrick told me that they have already seen a boost in productivity and developer happiness. Developers no longer need to wait for servers to be provisioned, and can now (as Jeff says) unleash their creativity and pursue their dreams. They expect to use Step Functions to improve the scalability, functionality, and reliability of their applications, going far beyond the initial use for the Coca-Cola Vending Pass. For example, Coke has built a serverless solution for publishing nutrition information to their food service partners using Lambda, Step Functions, and API Gateway.

Patrick and his team are now experimenting with machine learning and artificial intelligence. They built a prototype application to analyze a stream of photos from Instagram and extract trends in tastes and flavors. The application (built as a quick, one-day prototype) made use of Lambda, Amazon DynamoDB, Amazon API Gateway, and Amazon Rekognition and was, in Patrick’s words, a “big win and an enabler.”

In order to build serverless applications even more quickly, the development team has created an internal CI/CD reference architecture that builds on the Serverless Application Framework. The architecture includes a guided tour of Serverless and some boilerplate code to access internal services and assets. Patrick told me that this model allows them to easily scale promising projects from “a guy with a computer” to an entire development team.

Patrick will be on stage at AWS re:Invent next to my colleague Tim Bray. To meet them in person, be sure to attend SRV306 – State Machines in the Wild! How Customers Use AWS Step Functions.


Yes, Backblaze Just Ordered 100 Petabytes of Hard Drives

Post Syndicated from Andy Klein original https://www.backblaze.com/blog/400-petabytes-cloud-storage/

10 Petabyt vault, 100 Petabytes ordered, 400 Petabytes stored

Backblaze just ordered a 100 petabytes’ worth of hard drives, and yes, we’ll use nearly all of them in Q4. In fact, we’ll begin the process of sourcing the Q1 hard drive order in the next few weeks.

What are we doing with all those hard drives? Let’s take a look.

Our First 10 Petabyte Backblaze Vault

Ken clicked the submit button and 10 Petabytes of Backblaze Cloud Storage came online ready to accept customer data. Ken (aka the Pod Whisperer), is one of our Datacenter Operations Managers at Backblaze and with that one click, he activated Backblaze Vault 1093, which was built with 1,200 Seagate 10 TB drives (model: ST10000NM0086). After formatting and configuration of the disks, there is 10.12 Petabytes of free space remaining for customer data. Back in 2011, when Ken started at Backblaze, he was amazed that we had amassed as much as 10 Petabytes of data storage.

The Seagate 10 TB drives we deployed in vault 1093 are helium-filled drives. We had previously deployed 45 HGST 8 TB helium-filled drives where we learned one of the benefits of using helium drives — they consume less power than traditional air-filled drives. Here’s a quick comparison of the power consumption of several high-density drive models we deploy:

MFR Model Fill Size Idle (1) Operating (2)
Seagate ST8000DM002 Air 8 TB 7.2 watts 9.0 watts
Seagate ST8000NM0055 Air 8 TB 7.6 watts 8.6 watts
HGST HUH728080ALE600 Helium 8 TB 5.1 watts 7.4 watts
Seagate ST10000NM0086 Helium 10 TB 4.8 watts 8.6 watts
(1) Idle: Average Idle in watts as reported by the manufacturer.
(2) Operating: The maximum operational consumption in watts as reported by the manufacturer — typically for read operations.

I’d like 100 Petabytes of Hard Drives To Go, Please

“100 Petabytes should get us through Q4.” — Tim Nufire, Chief Cloud Officer, Backblaze

The 1,200 Seagate 10 TB drives are just the beginning. The next Backblaze Vault will be configured with 12 TB drives which will give us 12.2 petabytes of storage in one vault. We are currently building and adding two to three Backblaze Vaults a month to our cloud storage system, so we are going to need more drives. When we did all of our “drive math,” we decided to place an order for 100 petabytes of hard drives comprised of 10 and 12 TB models. Gleb, our CEO and occasional blogger, exhaled mightily as he signed the biggest purchase order in company history. Wait until he sees the one for Q1.

Enough drives for a 10 petabyte vault

400 Petabytes of Cloud Storage

When we added Backblaze Vault 1093, we crossed over 400 Petabytes of total available storage. For those of you keeping score at home, we reached 350 Petabytes about 3 months ago as you can see in the chart below.

Petabytes of data stored by Backblaze

Backblaze Vault Primer

All of the storage capacity we’ve added in the last two years has been on our Backblaze Vault architecture, with vault 1093 being the 60th one we have placed into service. Each Backblaze Vault is comprised of 20 Backblaze Storage Pods logically grouped together into one storage system. Today, each Storage Pod contains sixty 3 ½” hard drives, giving each vault 1,200 drives. Early vaults were built on Storage Pods with 45 hard drives, for a total of 900 drives in a vault.

A Backblaze Vault accepts data directly from an authenticated user. Each data blob (object, file, group of files) is divided into 20 shards (17 data shards and 3 parity shards) using our erasure coding library. Each of the 20 shards is stored on a different Storage Pod in the vault. At any given time, several vaults stand ready to receive data storage requests.

Drive Stats for the New Drives

In our Q3 2017 Drive Stats report, due out in late October, we’ll start reporting on the 10 TB drives we are adding. It looks like the 12 TB drives will come online in Q4. We’ll also get a better look at the 8 TB consumer and enterprise drives we’ve been following. Stay tuned.

Other Big Data Clouds

We have always been transparent here at Backblaze, including about how much data we store, how we store it, even how much it costs to do so. Very few others do the same. But, if you have information on how much data a company or organization stores in the cloud, let us know in the comments. Please include the source and make sure the data is not considered proprietary. If we get enough tidbits we’ll publish a “big cloud” list.

The post Yes, Backblaze Just Ordered 100 Petabytes of Hard Drives appeared first on Backblaze Blog | Cloud Storage & Cloud Backup.

Много ли са осем секунди

Post Syndicated from nellyo original https://nellyo.wordpress.com/2017/10/05/%D0%BC%D0%BD%D0%BE%D0%B3%D0%BE-%D0%BB%D0%B8-%D1%81%D0%B0-%D0%BE%D1%81%D0%B5%D0%BC-%D1%81%D0%B5%D0%BA%D1%83%D0%BD%D0%B4%D0%B8/

Зависи за какво.

Едно решение на съд в Лондон по делото England and Wales Cricket Board and Sky UK Limited vs Tixdaq Limited and Fanatix Limited  може да представлява интересзабележително е с извода, че осем секунди   съставляват съществена част от телевизионно предаване.

Става въпрос за отразяване на мачове по крикет. Самите предавания са  с времетраене около два часа. Според съда количествено не е налице използване на голяма част, но  от качествена гледна точка приложението дава достъп до най-важните моменти на мачовете – поради което има използване на съществена част от предаването, засяга се икономическия интерес на носителите на права.


Filed under: Media Law