DRM - Daniel’s Random Mutterings

I just got a new machine, with x64 processor (Pentium D Dual Core — nice machine!) and naturally it’s already running Linux. SUSE 10, to be specific.
Now I wanted to move my Django project over from the old machine and the Django installation kept failing. The setup.py told me:

Installed /home/tietze/temp/Django-0.91
Because this distribution was installed --multi-version or --install-dir,
before you can import modules from this package in an application, you
will need to 'import pkg_resources' and then use a 'require()' call
similar to one of these examples, in order to select the desired version:
    pkg_resources.require("Django")  # latest installed version
    pkg_resources.require("Django==0.91")  # this exact version
    pkg_resources.require("Django>=0.91")  # this version or higher

But calling django-admin.py always resulted in

No module named pkg_resources

– no mater how often I ran ez_setup.py or whatever. This was rather frustrating, since the Django installations on my previous machine were such a breeze.

It appears that the Python packaging tool is confused by the fact that 64 bit SUSE puts Python stuff under /usr/lib64 instead of /usr/lib and then assumes that it was installed to a non-standard location which isn’t on any Python path.

I finally found the solution here:
http://www.mail-archive.com/distutils-sig@python.org/msg01667.html

Specifically this recommendation:

“They need to add:

[easy_install]
site_dirs = /usr/lib64/python2.4/site-packages

to /usr/lib64/python2.4/distutils/distutils.cfg. This file probably
doesn’t exist; just create it.”

That did it. Created file, re-ran python ez_setup.py -U setuptools and then python setup.py install and now my new Django installation seems to be working. After a couple of minutes of frustration, that was actually quite easy.

We’re all probably part of several communities (portals, what-have-you) and follow information and news from multiple sources. Some we regularly contribute to, some we follow irregularly. Some we use for work, some for play, some for hobbies.
Me, I use, read, browse, and/or follow at least the following (off the top of my head) - Slashdot, Digg, Ebay, Amazon, Tauschticket, Golem.de, BoingBoing, OpenBC, Spiegel.de, heise.de, Stern.de, Joel On Software, Wikipedia.
Some of them I check for shopping, some for contacts, some for news, others for entertainment. But it’s all about my interests and my focus (OK, maybe not that focussed). What I haven’t yet found is a way to integrate and relate these different information sources in a way that is meaningful and helpful to me. (more…)

The next blushing bride for a Google wedding is waiting in the aisle: iRows - a Web-based spreadsheet application. She’s decked out in something old, something new, something borrowed and something Web-2 (point oh). Just perfect for adding to Google’s previous acquisition of Writely for a Google Web Office suite (GOffice?).

This really is the new bubble. Count me in!

Technorati Tags: Google, iRows, Web2.0

I spent this weekend working on a new portal site for which I had the idea a couple of days ago (yeah, I know — weekends are for something else; but it was absolutely freezing cold here in Germany, so I succumbed to my inner geek). I’m building the site using the Python Web framework Django and I am absolutely amazed at how productive that feels.

I’ve completed most of the data model and built several pilot pieces of functionality, such as logging in and out, registering as a new user, maintaining a personal user profile, uploading pictures, a personal homepage and personalized albums with all the images a specific user has uploaded to the site — along with dynamic thumbnails in any given size. I know that there’s still a lot of ground to cover. Especially in the area of fool-proofing and tightening the whole thing; testing for and catching all kinds of bad errors. But then the net result is really astounding for roughly one-and-a-half days of work. Bear in mind, I’m learning Python and Django while I go along. I’m usually a Java guy.

Hats off to the Django guys!

This seems to be quickly becoming a new and viable business model: Start small, do something really cool, build a loyal user base, and then be acquired by Google or Yahoo! or Microsoft. Sounds like a good summary for the “exit strategy” section of the business plan you present to VCs.

Most recent success: Writely — purchased by Google. Previously: flickr, Del.icio.us (now part of Yahoo!) and many more in the past weeks and months.

The problem, obviously, is that this business success is not predictable. And that it’s bloody hard to come up with something slick and captivating and to build up a following, while others are most likely doing the same thing at the same time. Otherwise everybody would be doing it. This form of exit from a startup can be very exciting for the founders and staff. If I had a cool idea, I’d hop on the bandwagon in an instant.

Do we know of anybody who’s been actively pursuing this strategy from the get-go? Figure out what might be a strategic advantage for Google, Yahoo, MSFT in the future, develop the idea independently and make enough noise so you are noticed by your prospective future owner? Or is it always serendipity?

Technorati Tags: Google, Writely, Business

After an initial pointer from Bob Walsh, over at the Joel On Software discussion board, and a follow-up blog posting from him I recently went and checked out Squidoo and played around with it a bit. In case you haven’t heard of it yet, Squidoo is a knowledge sharing portal which allows “lensmasters” to create “lenses” — pages of information about things they know about (or claim to know about). Squidoo’s tagline is “Everybody’s an expert on something”. I put together my own lens, on something think I know about, and fiddled around with the modules.

Somehow, I was extremely underwhelmed. The site editing, inside the Web browser, AJAX-style, feels sluggish and unresponsive and not “fun” enough. I don’t feel that I have much control over the site, page layout, page contents (just as a small f’rinstance, I added a picture to my lens, which shows up fine in the editing view but is simply not there in the published lens). The available modules that can be arranged in the lens and populated with content based on the module type (e.g. Amazon links for the Amazon module) all seem more or less geared towards traffic generation and revenue generation (think AdWords and Amazon Partner Network) rather than true information sharing. As a result, the lenses I’ve looked at don’t have much of a ‘networked’ feel to them. This, coupled with the fact that the whole site itself feels too sluggish and slow doesn’t really tempt me to explore. Which brings me to my original question: WHY would I, as a viewer, want to go to Squidoo to look for stuff? And if there is no viewing audience, why bother to post there in the first place?

Yes!!! One of my favourite cartoon characters is back. Berkeley Breathed’s Opus the penguin, of Bloom County fame, reappeared a few weeks ago on comics.com, with a weekly strip. Berkeley had announced this some time ago, and now we can read it online.
If you can, grab the Bloom County cartoon collections from Amazon. You won’t regret it.
And now repeat after me - ‘Pear Pimples for Hairy Fishnuts’.

Technorati Tags: Cartoons, Opus

I used to like Amazon’s recommendation feature, with recommendations based on my purchase history. I actually used it regularly to browse the Amazon site, looking for new and interesting stuff. Quite often, I’d end up ordering a couple of the recommended books. Very nice!
Recently, though, they’ve changed something in the system and now I hate it. Now, recommendations are based on stuff you look at, or you’ve looked at in the past. This sounds like an idea a computer guy would come up with and quickly implement, over the weekend. In reality, it’s terrible. You just look at one book about veruccas — not because you’re actually interested, but because someone you sorta kinda know mentioned this book as a possible birthday present and you’re checking to see how much it costs — suddenly your whole recommendation list and “personal page” is full of unpleasant-looking books about weird foot diseases (with a foot-fetish softporn book thrown in for good measure).
That’s not a recommendation. Thats like an extremely pushy bazaar stall owner trying to aggressively sell me every single thing I happen to look at. Also, it seems that every time I look they’ve moved the button for clearing my viewing history into a new hard-to-find place in the menu.

Amazon - give me back my recommendations. I want to spend my money!

Technorati Tags: Amazon,Books

No, not needs. Want. I *want* one

I don’t know, why. I don’t really like Macintoshes. In fact, I sneer at Macs and their puny one-button mice. Rationally, I think I’d be much better off getting a decent PC and installing Linux on that. At least for application availability, as well as interoperability with my other machines, I’m pretty sure that would be the better approach. I don’t really need a third operating system. Also, the Laptop PC I’d get for the dough a MacBook Pro costs would simply blow me away.
But still — I want one. It just looks so sleek and well-designed. And it holds the promise of all the pieces being made for each other (as opposed to the hassle of getting a Linux WiFi driver to work on the laptop which has a WiFi interface chip you don’t know, as well as an unsupported video card).

Help me. Please.

Technorati Tags: Tech