Websites in review – Week 1

I often come across new websites, quite often by accident, or by indirection in links from looking at other details. The Internet is an amazing place, and one could spend all day reading such a variety information and only touch on just a few specific topics.

I think it’s important to share interesting and new sites, often it’s a referral from others that provide for enjoyable and useful reading. Here are mine in return. Sites which I bookmarked and intended to review again.

99 Designs —

Need something designed? 99designs connects clients needing design work such as logo designs, business cards or web sites to a thriving community of 17,781 talented designers.

Wufoo —

Wufoo strives to be the easiest way to collect information over the Internet.

Our HTML form builder helps you create contact forms, online surveys, and invitations so you can collect the data, registrations and online payments you need without writing a single line of code.

Mako Templates —

Mako is a template library written in Python. It provides a familiar, non-XML syntax which compiles into Python modules for maximum performance. Mako’s syntax and API borrows from the best ideas of many others, including Django templates, Cheetah, Myghty, and Genshi. Conceptually, Mako is an embedded Python (i.e. Python Server Page) language, which refines the familiar ideas of componentized layout and inheritance to produce one of the most straightforward and flexible models available, while also maintaining close ties to Python calling and scoping semantics.

Venture Beat —

I came across this site when reading Developer Analytics: Facebook game Mob Wars making $22,000 a day.

VentureBeat’s mission is to provide news and information about private companies and the venture capital that fuels them. Founder Matt Marshall covered venture capital for the San Jose Mercury News until he left in Sept. 2006 to launch VentureBeat as an independent company. VentureBeat will focus initially on Silicon Valley, and gradually expand to cover innovation hubs around the globe. Its mission in each region will be the same: to provide insider news and data about the entrepreneurial and venture community that is useful to decision makers.

Developer Analytics —

The previous article referenced this site, with the tagline “the world’s first social media ratings and reporting services.” I like the categories used in reports.

  • Reach – Unique traffic.
  • Audience Profile – Demographics and sociographics breakdown. Interest clouds.
  • Engagement -Impressions. Average page views per user. Return users.
  • Growth -New Users. Churn Rates. Viral Factors.
  • Monetization – CPM and CPA revenue potential.

SXSW Interactive —

I’ve heard of South By South West Conference before, and this is it. SXSW Interactive: March 13-17, 2009 -The Brightest Minds in Emerging Technology

SXSW Interactive The SXSW Interactive Festival features five days of exciting panel content and amazing parties. Attracting digital creatives as well as visionary technology entrepreneurs, the event celebrates the best minds and the brightest personalities of emerging technology. Whether you are a hard-core geek, a dedicated content creator, a new media entrepreneur, or just someone who likes being around an extremely creative community, SXSW Interactive is for you!

Actually sounds rather interesting, and different.

Pulse 2.0 —

Pulse 2.0 is a company that is driven by our passion of technology and entrepreneurship. We use Pulse 2.0 to share our thoughts on Web 2.0.

Monetization through Online Advertising

Next week I will be the panel moderator for September 2008 Entrepreneurs Forum on “Monetization through Online Advertising” organized by Ultra Light Startups™. A slightly different approach to my regular speaking schedule, it will be good to observe and interact with our speakers.

The companies represented by the panelists for the evening include:

  • PerformLine – Alex Baydin, Founder and CEO
  • SocialDough – Derek Lee, Founder and CEO
  • Zanox Inc – Rani Nagpal, Vice President of Affiliate Management
  • Blinkx – Max Ramirez, Head of online ad sales

Event Details

Date: Thursday, September 4, 2008
Time: 7:00pm – 9:00pm
Location: Rose Technology Ventures, LLC
Street: 30 East 23rd Street
City/Town: New York, NY

A 5.1 QEP nicety – Using join buffer

I was surprised to find yesterday when using MySQL 5.1.26-rc with a client I’m recommending 5.1 to, some information not seen in the EXPLAIN plan before while reviewing SQL Statements.

Using join buffer

| id | select_type | table | type   | possible_keys | key          | key_len | ref                    | rows  | Extra                                        |
|  1 | SIMPLE      | lr    | ALL    | NULL          | NULL         | NULL    | NULL                   |  1084 | Using where; Using temporary; Using filesort |
|  1 | SIMPLE      | ca    | ref    | update_check  | update_check | 4       | XXXXXXXXXXXXXXXXX      |     4 | Using where; Using index                     |
|  1 | SIMPLE      | ce    | ALL    | NULL          | NULL         | NULL    | NULL                   | 13319 | Using where; Using join buffer               |
|  1 | SIMPLE      | co    | eq_ref | PRIMARY       | PRIMARY      | 4       | XXXXXXXXXXXXXXXXX      |     1 | Using where                                  |
4 rows in set (0.00 sec)
mysql> select version();
| version() |
| 5.1.26-rc |
1 row in set (0.00 sec)

Sergey Petrunia of the MySQL Optimizer team writes about this in Use of join buffer is now visible in EXPLAIN.

An intestesting approach to free hosting

I came across the OStatic Free hosting service that provide Solaris + Glassfish (Java Container) + MySQL.

They offer “Now you can get free Web hosting on Cloud Computing environment free of charge for up to 12 months.

The catch “accumulate 400 Points for participating on the site“.

A rather novel approach, you get 100 points for registering, but then only 5-15 points per task. The equates to approximately at least 30 tasks you need to perform, such as answering a question, or reviewing somebodies application. That seems like a lot of work?

In this case, Free has definitely a cost and time factor to consider.

Get linked to Drizzle

We are always looking at different ways to help promote, inform and identify contributers, users and supports for Drizzle.

One way is to join the Linked In Drizzle group (click here when logged in). You will already see a few MySQL die hards, but also the need breed of names that are part of Drizzlemania.

VirtualBox, compiling Part 2

So I managed to find all dependencies after some trial and error for compiling VirtualBox 1.6.4 under Ubuntu 8.0.4, then finding the Linux build instructions to confirm.

It was not successful however in building, throwing the following error:

kBuild: Compiling dyngen - dyngen.c
kBuild: Linking dyngen
kmk[2]: Leaving directory `/usr/local/VirtualBox-1.6.4/src/recompiler'
kmk[2]: Entering directory `/usr/local/VirtualBox-1.6.4/src/apps'
kmk[2]: pass_bldprogs: No such file or directory
kmk[2]: *** No rule to make target `pass_bldprogs'. Stop.
kmk[2]: Leaving directory `/usr/local/VirtualBox-1.6.4/src/apps'
kmk[1]: *** [pass_bldprogs_before] Error 2
kmk[1]: Leaving directory `/usr/local/Virtu

More searching, I needed to add two more files manually. Read More Here.

A long wait, compiling for 20+ minutes, and a necessary reboot as upgraded images threw another error, I got 1.6.4 running, and able to boot Fedora Core 9 image created under 1.5.6

But the real test, and the need for this version was to install Intrepid.

This also failed with a Kernel panic during boot. More info to see this reported as a Ubuntu Bug and Virtual Box Bug.

More work still needed.

Interacting with BuildBot using IRC

Using BuildBot for Drizzle has been a great way to help in the verification of the sometimes rapid code changes that are being committed.

Curious why the IRC notifier within BuildBot only joined and exited the #drizzle channel in IRC, some further investigation of the IRC Documentation lead to more information to share.

By default, the following configuration is not much help in any automated notification.

from buildbot.status import words
c['status'].append(words.IRC(host="", nick="drizzle_buildbot", channels=["#drizzle"]))

However, within IRC you can query using several commands. My first trials.

rbradfor: drizzle_buildbot: list builders
[3:10pm] drizzle_buildbot: Configured builders: centos5.64.1 centos5.64.1-mt debian4.32.1[offline] debian5.32.1 debian5.32.2 debian5.64.1 doxygen fedora8.32.1[offline] fedora8.64.1 gentoo8.32.1 gentoo8.64.1 osx105.32.1 osx105.32.1-mt osx105.64.1[offline] osx105.64.1-mt[offline] suse11.32.1[offline] ubuntu804.32.1[offline] ubuntu804.32.2[offline] ubuntu804.32.3[offline] ubuntu804.32.4 ubuntu804.32.4-mt ubuntu804.32.5 ubuntu804.32.6[offline] ubuntu804.32.7[offline] ubuntu804
[3:10pm] rbradfor: drizzle_buildbot: status all
[3:10pm] drizzle_buildbot left the chat room. (Excess Flood)
[3:11pm] drizzle_buildbot joined the chat room.
[3:11pm] rbradfor: drizzle_buildbot: notify on
[3:11pm] drizzle_buildbot: The following events are being notified: ['started', 'finished']
[3:13pm] drizzle_buildbot: build #484 of centos5.64.1 started including []
[3:18pm] drizzle_buildbot: build #484 of centos5.64.1 is complete: Success [build successful]  Build details are at
[3:25pm] rbradfor: drizzle_buildbot: notify off
[3:25pm] drizzle_buildbot: The following events are being notified: []
[3:26pm] rbradfor: drizzle_buildbot: watch centos5.64.1
[3:26pm] drizzle_buildbot: there are no builds currently running
[3:34pm] rbradfor: drizzle_buildbot: notify on failed
[3:34pm] drizzle_buildbot: The following events are being notified: ['failed']
[4:09pm] rbradfor: drizzle_buildbot: help
[4:09pm] drizzle_buildbot: Get help on what? (try 'help foo', or 'commands' for a command list)
[4:09pm] rbradfor: drizzle_buildbot: help commands
[4:09pm] drizzle_buildbot: Usage: commands - List available commands
[4:09pm] rbradfor: drizzle_buildbot: commands
[4:09pm] drizzle_buildbot: buildbot commands: commands, dance, destroy, excited, force, hello, help, join, last, leave, list, notify, source, status, stop, version, watch

The docs list the following commands.

To use the service, you address messages at the buildbot, either normally (botnickname: status) or with private messages (/msg botnickname status). The buildbot will respond in kind.

Some of the commands currently available:

list builders
    Emit a list of all configured builders
status BUILDER
    Announce the status of a specific Builder: what it is doing right now.
status all
    Announce the status of all Builders
    If the given Builder is currently running, wait until the Build is finished and then announce the results.
    Return the results of the last build to run on the given Builder.
    Join the given IRC channel
    Leave the given IRC channel
notify on|off|list EVENT
    Report events relating to builds. If the command is issued as a private message, then the report will be sent back as a private message to the user who issued the command. Otherwise, the report will be sent to the channel. Available events to be notified are:

        A build has started
        A build has finished
        A build finished successfully
        A build failed
        A build generated and exception
        The previous build was successful, but this one failed
        The previous build failed, but this one was successful

    Describe a command. Use help commands to get a list of known commands.
    Announce the URL of the Buildbot's home page.
    Announce the version of this Buildbot.

If the allowForce=True option was used, some addtional commands will be available:

force build BUILDER REASON
    Tell the given Builder to start a build of the latest code. The user requesting the build and REASON are recorded in the Build status. The buildbot will announce the build's status when it finishes.
    Terminate any running build in the given Builder. REASON will be added to the build status to explain why it was stopped. You might use this if you committed a bug, corrected it right away, and don't want to wait for the first build (which is destined to fail) to complete before starting the second (hopefully fixed) build.

I don’ want to flood the IRC channel with messages, so delving deeper into the documentation via the following commands gives me more tips.

$ cd buildbot-0.7.8
$ pydoc buildbot.status.words

By defining categories against the IRC notification, and assigning builders to a given category, in theory you will get notifications just for these builders. I didn’t seem to produce the desired results, so for now it needs to be manual interaction until I get additional time to investigate.

b00 = {'name': "centos5.64.1", 'slavename': "centos5_64", 'builddir': "build00", 'factory': f1, 'category': "irc" }
from buildbot.status import words
c['status'].append(words.IRC(host="", nick="drizzle_buildbot", channels=["#drizzle"], categories=["irc"]))

Virtual Box, a world of hurt

I successfully installed Virtual box via a few simply apt-get commands under Ubuntu 8.04 via these instructions.

It started fine, after two small annoying, install this module, add this group messages. I was even able to install Ubuntu Intrepid from .iso. But from here it was down hill.

Attempting to start VM gives the error.

This kernel requires the following features not present on the CPU:
Unable to boot - please use a kernel appropriate for the CPU

Some digging around, and confirmation that the current packaged version of Virtual Box doesn’t support PAE. You think they could tell you before successfully installing an OS. I’m running 1.5.6, I need 1.6.x

$ dpkg -l | grep virtualbox
ii  virtualbox-ose                             1.5.6-dfsg-6ubuntu1                      x86 virtualization solution - binaries
ii  virtualbox-ose-modules-2.6.24-19-generic   24.0.4                                   virtualbox-ose module for linux-image-2.6.24
ii  virtualbox-ose-source                      1.5.6-dfsg-6ubuntu1                      x86 virtualization solution - kernel module

Off to the Virtual Box Downloads to get 1.6.4
Don’t make the same mistake as I did and use the first download link, that’s the commercial version that doesn’t install what you expect, you need the OSE. Of course this is not packaged, it’s only source.

Checking for environment: Determined build machine: linux.x86, target machine: linux.x86, OK.
Checking for kBuild: found, OK.
Checking for gcc: found version 4.2.3, OK.
Checking for as86:
  ** as86 (variable AS86) not found!

Ok, well I go through this step like 4 times, installing one package at a time, I wish they could do a pre-check and give you all missing requirements. I installed bin86, bcc, iasl.

Then I got to the following error.

$ ./configure
Checking for libxml2:
  ** not found!

Well it’s installed, all too hard. Throw Virtual Box away for virtualization software. And why am I using it anyway. Because VMWare Server doesn’t work under Ubuntu 8.04 either because of some ancient gcc dependency. Sees I may have to go back to that. I just want a working virtualization people on the most popular Linux distro to install other current distros. It’s not a difficult request.

$ dpkg -l | grep libxml
ii  libxml-parser-perl                         2.34-4.3                                 Perl module for parsing XML files
ii  libxml-twig-perl                           1:3.32-1                                 Perl module for processing huge XML document
ii  libxml2                                    2.6.31.dfsg-2ubuntu1                     GNOME XML library
ii  libxml2-utils                              2.6.31.dfsg-2ubuntu1                     XML utilities
ii  python-libxml2                             2.6.31.dfsg-2ubuntu1                     Python bindings for the GNOME XML library

Choosing MySQL 5.1 over 5.0

I have been asked twice this week what version of MySQL I would choose for a new project.
As with most questions in life the answer is: It Depends?

In general I would now recommend for a new project to select 5.1, and he is why.

  1. If it’s a new project and your not managing existing applications with older versions then 5.1 is slated for General Availability (GA) at some imminent time. Having been at Release Candidate (RC) for quite some time (almost 1 year), many people, both internally and in the community are just waiting for Sun/MySQL to get this version out.
  2. MySQL 5.0 is in maintenance mode, it’s now 3 years old. MySQL is placing (I’m assuming) resourcing energies to current and future releases.
  3. If your looking at releasing a product in the next 3 months for example, you do not want to consider the testing and deployment of a new version (e.g. 5.1) in the next 6-9 months.
  4. Unless your comparing specific performance between 5.1 and 5.0 in your edge cases, for a new project start with 5.1 you should be testing and confirming performance and reliability here. The worse case is you can test in 5.0 of any specific problem.
  5. 5.1 gives you new features of course, partitioning may be of benefit but don’t assume it’s going to be a great improvement unless you applications SQL naturally tended to the MySQL partitioning strengths.
  6. The single biggest benefit is the Pluggable Storage Engine Architecture. This can give you some benefits, and in the case of transaction storage engines that are production ready, Innodb now has a pluggable version, much improved on the MySQL supplied version. There are a long list of other engines under development with relative strengths and weakness, however be wary of versions that require customized builds of MySQL.

There are some concerns where I don’t have answers? For example, if you have MySQL Support , is 5.1 supported? I know a common answer to problems in pre 5.0 versions is, have you tried upgrading to 5.1

Why is not released? This is good question, the answer is obviously a level of quality, however it is generally discussed that 5.1 is of better quality in existing features then 5.0. It is 5.1 specific features you need to be careful of. It’s important that you do read carefully the 5.1 Release Notes to see where bug fixes or compatibility changes are still occuring.

As with any choice in the Open Source world, some level of risk assessment is necessary. If you have good metrics and measurement in place for your system, and you adequately test your software, there is no reason not to now consider 5.1 as a viable alternative for new development.

As I post this I note, I see the yet unreleased 5.1.28 list of bugs still shows issues of concern.

Project Darkstar

It may sound like either a astronomical research project or a Star Wars spin- off, but Project Darkstar is an open source infrastructure from Sun Microsystems that states “simplify the development and operation of massively scalable online games, virtual worlds, and social networking applications.”

The advertising sounds promising like many sites, the emphasis seems to be on gaming throughout the material, interesting they threw in the term “social networking applications” specifically in opening descriptions.

I believe worthy of investigation, if only to see how that solve some classic problems. So, Learn some more, Start your rockets and Participate.

Finding, exposing and referencing good material

I came across by accident. Like many sites and information these days, you simply don’t find via search engines because your normally searching for something specific. I did find it via several levels of hyperlinks. I really wish there was firefox plugin that would track every site you visited, often I’d like to plot how I got to where I am, but that’s another story.

Considering the author is Australian, a top Plurker and Photographer got me intrigued enough to delve for a few moments. What I found is some good information, such as 10 Ways to Optimize a Popular Post on Your Blog and Is Your Blog a Networking Tool?.

Some more reading, Five Ways That Strategic Bullet Points Make You a Stronger Blogger, which leads to a site that includes articles such as Seal the Deal: 10 Tips for Writing the Ultimate Landing Page, which is exactly what I’m looking for with an upcoming Ad Words campaign, but wasn’t searching at the time.

You never know sometimes where good information comes from.

Ultra light startups NY meeting

I attended the Ultra light startup’s meeting last night for the first time. I found it most productive for the 2 hours of time to see a different approach talking about startups, to see a variety of approaches, concepts, ideas, ventures all at various stages and generally people with different and interesting ideas and goal.

The start included a 1-2 min elevator pitch by every person with a few questions of feedback. Some interesting projects included, Rose Tech Ventures Incubator, Home Shop Technologies, New York City Co-working, Wiki Streets, Robots for Planet Earth, Festival Travel Channel, Sunshine Suites, Peek You and Wiki Pages. Two presenters put forth their ideas,concepts, and intentions with domains registered within the last 2 days.

The main discussion was on Co-Working, a concept I’ve not heard of before. It’s a different approach to the Telecommuting approach, companies moving from attendance based to performance based. Another term mentioned as ROW – Results Only Work environment.

I think some improvements in the “elevator pitch” would be.

  • egg timer for 90 seconds
  • Recommend people have 3 slides, and example layout would be “How I am”, “What we do”, and “What we want”
  • Some tips of not what to put on slides, like for example, more then 5 lines and less then 30 point for example.

Facebook Group
Subscribe to mailing list

Using consistent data types for columns

I came across this error recently when trying to modify the data type of a column.

ERROR 1025 (HY000): Error on rename of './sakila/#sql-1d91_5' to './sakila/inventory' (errno: 150)

Not the first time, and not the last time. A common problem with InnoDB tables, is the lack of information, you need to dig deeper with the following command (and appropriate security a well organized security profile will NOT have).



080717 20:00:28 Error in foreign key constraint of table sakila/inventory:
there is no index in the table which would contain
the columns as the first columns, or the data types in the
table do not match the ones in the referenced table
or one of the ON ... SET NULL columns is declared NOT NULL. Constraint:
  CONSTRAINT "fk_inventory_film" FOREIGN KEY ("film_id") REFERENCES "film" ("film_id") ON UPDATE CASCADE


You also need to dig though the output of the command to find this, on a larger system this can be quite a lot of information just to find the details of the error. (It would be nice if there was an easier way.)

The result of this error is the columns in a foreign key relationship need to be of the same data type. This is actually a good thing, and MySQL generally operates much better in joins when the join columns are consistent.

On the assumption that you use surrogate primary keys for tables, this is a candidate for a naming standard for primary keys should be the primary key name is unique column name within your schema.

For example, if you call all your primary key’s ‘id’, your foreign key’s are normally ‘table_id’. While this is a common approach that I promoted myself for many years, it’s easy to read and consistent, actually naming every primary key uniquely provides two great benefits.

First, you can easily identify relationships in your entire schema without even knowing about the schema in detail. Second, you can leverage the benefit of the INFORMATION_SCHEMA to in the case of this post, confirm the data types are consistent for all matching columns, even when Referential Integrity is not used.

So, instead of using ‘id’, use ‘actor_id’, ‘film_id’, ‘user_id’ etc. For any self join keys, I normally prefix with parent, so ‘parent_user_id’ for example if you have a hierarchy within a table.

Extending application data to the cloud

I was one of the invited panel speakers to A panel on Cloud Computing this week in New York. As one of 2 non vendor presenters, it was a great experience to be invited and be involved with vendors.

While I never got to use my slides available here, I did get to both present certain content, and indeed questions and discussions on the night were on other points of my content.

Cloud computing is here, it’s early days and new players will continue to emerge. For example, from the panel there was AppNexus, reviewed favorably at Info World in comparison with EC2 and Google App Engine, 10gen, an open source stack solution and Kaavo which from an initial 60 seconds of playing provide a management service on top of AWS similar to what ElasticFox provides. I need to investigate further how much the feature set extends and would compete with others like RightScale for example.

The greatest mystery came from Hank Williams and his stealth Kloudshare. He did elaborate more on where they aim to provide services. A new term discussed was “Tools as a service”, akin to moving use metaphorically from writing in Assembly language to the advanced frameworks of today’s generation of languages such as Java and Ruby.

Thanks to Murat Aktihanoglu of Unype who chaired the event.

MySQL involvement in OSCON opening keynote

Before I get to post my OSCON reflection I see I didn’t post this (which I reference).

At OSCON opening keynotes Tim O’Reilly Interviews Monty Widenius & Brian Aker. This provided some interesting answers in a Q & A session. Here is some of the discussion.

TO: So 6 months in. How is it with Sun?
BA: Really rewarding environment. My first question was? You are going to send me free H/W. No H/W has been delivered yet, or access to the masses, still hoping. Sun is a very engineering driven company.
MW. Thanks God we didn’t go public. Starting to do closed sourced components, going public this would have continued.

TO: Sun saved MySQL from public market/ insulated from market.
MW: 6 months in, Sun still trying to figure out what they bought. Sun has made a commitment to open source throughout the organization. Engineers who have been working in closed environments, now seeing this all in public, and opens yourself up to more inputs and exposure.

TO: You have your own projects within sun, how does that affect with the main line of development of MySQL, Monty you with the Maria Storage Engine, Brian you with Drizzle.

TO: What is the Support like in Sun?
BA: My boss got it. We are looking at going after different market area, niche and ecosystem. There is certain direction the main codebase is heading such as enterprise features, oracle like replacements. There is a core set of environments what they aren’t needed. Additional new requirements like a proximity data storage, historically Postgres has been good for this type of GIS data. This is a new type of data store. location/time and proximity of objects.
Sun has given us more free hands to work for best features of MySQL. For Drizzle, to strip it down into more components architecture and extensibility. It’s a micro-kernel there will be an interface for large parts of the code.

TO: What do you think about Google?
BA: Happy opening up more of their data, and trying to turn the world into their own 20% of time project.

TO: What do you think about Amazon?
BA: Interesting position, secretive company. At the beginning how little anybody though of Amazon in a service marketplace.

TO : What do you think about Microsoft?
MW: less and less things are good.
BA: Irrelevant.

TO: What do you think about Apple?
NW: More afraid of Apple then Microsoft
BA: Really want an iPhone, but hoping Google will get Android out and it works.

TO: What are the cool things MySQL can do on the Sun field, and reverse?
BA: Both Sun and MySQL Engineers thought about open source differently. MySQL has created a set of steps of evolution, e.g. employees contributing to open source projects. MySQL’s DNA was very small, it’s interesting how fast this is influencing Sun’s approach
MW: MySQL has become to management driven in previous years, Sun has enabled us to get back to our roots.