January 27, 2010

iPad Name Already in Use

One thing that Apple appears to have lost sight of in all the thrills of announcing their new tablet computer is that Widgetaria, presumably an Apple business partner, already promotes software called iPad. I wonder if they care?

January 17, 2010

Welcome BlueBream

Finally, a chance to start afresh with a promising set of technologies that has consistently been under-utilized. I am not even going to mention the old  name of the project, simply say that if you use Python you have likely heard of it. I have blogged  in the past about the confusion of search results.  Hopefully under the new name the vast preponderance of search results will be accurate and useful. So go meet BlueBream.

So, One of Life's Little Frustrations ...

... is the way Vista insists on taking the numbers in my filenames as numbers, thereby sorting them in precisely no convenient order. These files, however, are effectively named using a Hamming Code, so 62 is actually greater than 212, just like XP always knew. I just know this is going to turn out to be one of those awful tweaks that I should have put in three minutes after starting Vista.


I don't think we'll bother those nice Python people with that one. Which probably dooms this post to have me as the single reader.

Sudden pleasant thought. I will email the client to whom I first promised this work and then they will know that work is not only underway but almost complete. I have to finish this today or send an incomplete deliverable, since the client has paid for the classes, which have been delivered.

Happy Sunday. And people think that consulting is a part-time job ...

Not Really Python, But ...

We interrupt the newly established Python-only regime with a message from your fellow human-beings in Haiti.

if YouGiveAShit:

'Nuff said. This is a long-term project.

January 16, 2010

Definitely Not P*thon

And, to  complete my test of the feed reconfiguration I am going to post something that isn't about the P-language and ensure that it doesn't show up in Planet P*thon.

To make ad hominem remarks about the collective denizens of the planet would be like talking behind their backs - and besides, what if the test fails?

January 15, 2010

Just the Python, Please

The webmasters are having a spring clean and suggested that I might like to provide a "pure Python" feed. Michael Foord pointed out that such feeds existed already. If you want to skip the irrelevant waffling and occasional foray into politics then from now on Planet Python should see such a feed.

Direct subscribers wanting only Python can follow
    http://holdenweb.blogspot.com/search/label/python
    http://holdenweb.blogspot.com/feeds/posts/default/-/python
if they prefer.

Other blogs:
    A Yorkshireman in America where the rabid politics really live
    On Your Desktop lame celebrity desktop blog
    PyCon conferences - you probably get this via the Planet
    Python Software Foundation News others contribute more than I do

January 7, 2010

Absent DB API Test Files Fixed

I saw a somewhat annoying comment from a web site reader recently:
The following message was received via the web site.

From:       
Email:      
Subject:    /PyConTX2007/dbapi.tgz  Not Available
Telephone: 

I am grateful to the anonymous commenter for letting me know that I'd failed to redirect certain essential links in the recent web site migration.

It would have been even better if I could have replied to the comment to let my reader know that the problem is fixed (which it now appears to be). Alas he or she left absolutely no contact information whatsoever.So all I can do is record the fix and hope that this file wasn't being urgently sought (which I doubt it was).

January 6, 2010

Register for PyCon, or the Kitten Gets It



Just sayin', it might be better if you planned to attend the conference.

It's not that I like harming innocent small furry creatures, it's just that there are still thousands of Python users and potential Python users who still don't know what excellent value for money the conference is, or how much fun you can have there.

PyCon is in Atlanta this year, and despite the parlous state of the world economy there's a chance that it will be the biggest Python event ever. But hey, we all know that size isn't everything.

Although I was notionally on the program committee, and did at least register my opinions of the talks I was allocated by the submissions scheme, I (yet again) didn't manage to make a single meeting due to pressure of other business.  I'm actually not too unhappy about that, given the incredibly difficult task that the committee had to perform in order to ... well, I was going to say "sort the wheat from the chaff", but in fact there was so little chaff that wouldn't really be an appropriate analogy.

A number of people have blogged already about Five PyCon Talks You Must Not Miss, but since there's still a few hours to get in at the early bird rate I thought I'd throw out my list of unmissables from the extensive list of talks. How's that for arrogance?

A Short Pinax Tutorial, Danny Greenfeld. I have heard Danny speak at local Python user groups. If I had time I would be attending the half day Pinax tutorial that he and James Tauber are giving, but this is the second best way to find out how to get started with Pinax.

What Every Developer Should Know About Database Scalability, Jonathan Ellis. This one will be straight from the horse's mouth - it's always worth hearing Jonathan's summaries of his immense practical experience.

IronPython Tooling, Dino E Viehland. As a Windows user who has hardly touched IronPython so far I am interested in finding out what my options are. Nobody is likely to know better than Dino, who is probably the most prominent member of the Microsoft development team.

Scaling Python Webapps from Zero to 50 Million Users - A Top-Down Approach, Jinal Jhaveri. Although I haven't heard Jinal speak before I can't resist the lure of hearing what someone with really high-volume Python web experience has to say.

Why Not Run All Your Tests All the Time? A Study of Continuous Integration Systems, Titus C Brown. Ever since I hear Titus' tutorial (with Grig Gheorghiu) on testing a couple of year's ago I've wanted to hear what he has to say about CI. This is my chance.

Alas, five talks isn't anywhere near enough to encompass everything I want to hear, and I am also keenly anticipating Turtles All The Way Down: Demystifying Deferreds, Decorators, and Declarations, Pay Only for What You Eat: A Tour of the Repoze.BFG Repository and Philosophy, On The Subject Of Source Code, Python's Dusty Corners, Debating 'til Dawn: Topics to Keep You Up All Night, Powerful Pythonic Patterns and Tests and Testability.  This is going to be a knockout conference even before you consider all the amazing things that will be happening in Open Space.

You really have to be there. And you'll be helping to save a poor kitten that never did anyone any harm in its life!

January 5, 2010

Two Days Left to be a PyCon Early Bird

Just a quick reminder that PyCon early bird registrations end on January 6, so you don't have long to take advantage of these rates. Quick! Before you forget!

January 2, 2010

Why Save MySQL Now?

I am having a hard time understanding why Michael "Monty" Widenius is pushing so hard to Save MySQL as the Oracle takeover of Sun Microsystems looms. It was Widenius himself, one of two primary developers of the product, who introduced a dual-licensing model that allowed clients to purchase supported products from his company, MySQL AB, or to download the open source components themselves.

After a considerable length of time running the company and presumably directing its operations he agreed to sell MySQL AB (including the rights to the MySQL branding and trademarks, apparently) to Sun Microsystems in January 2008, and went to work in Sun's MySQL division. One presumes that a good portion of the billion dollars that Sun paid for the company went to Widenius.

He then proceeded to undermine the MySQL 5.1 launch at the end of November the same year with a rant about its unsuitability for purpose (for which he strangely appeared to feel little or no responsibility) that sent a terrible message to the user community. Then he followed that up less than three months later with an announcement that he was leaving Sun and had started his own business, Monty Program AB. Fair enough.

He now apparently perceives a threat to the MySQL "brand" because Oracle will have no vested interest in ensuring the vigorous development of the MySQL software to compete with its high-margin brand. The question that I am left with is: why, having sold the brand (and since MySQL was open source licensed under version 2 of the GPL, the brand was the main intellectual property transferred along with the support business) does Monty imagine he has any right to control what happens to it?

The software itself is protected under the GPL, so a fork is perfectly practical (and indeed Monty's new company has for some time been working on a branch they call Maria). So all we are talking about is a name, and one which Monty willingly sold. What's the deal?