I wrote a short article on writing Excel files with Python and exporting them with Django on developer.com. The site caters to a (mostly) Java and .NET crowd, but they’re expanding into Python and Ruby. I’m going to try and write an article or two a month for them. I’ll link to them here. The topics will probably be a little bit basic for the regular Python crowd, the goal is to bring some new folks into the fold.
Delicious Links
- Expanding the hard drive space on VMWare Fusion [Archive] - Mac Forums
- easier to post to than a wiki: PyGTK, Py2exe, and Inno setup for single-file Windows installers
- Apache: Creating A Session-Aware Loadbalancer Using mod_proxy_balancer (Debian Etch) | HowtoForge - Linux Howtos and Tutorialsincludes how to set up sticky sessions, which are slightly tricky
- SDSS SkyServer DR7
- Shindig - an Apache incubator project for OpenSocial and gadgetsa reference implementation of a OpenSocial / gadget container
Latest Tastebud
- Chicago BBQ Styles on 848 This MorningA brief history of Chicago BBQ. Interesting stuff, they also interview Barry Sorkin, the owner of Smoque.
How could you improve a shark? Laser beams. Frick’n lasers!
But in all seriousness, and I realize this is a tangent to the point of your article, but it’s something we’ve struggled with in our research work, anyway, spreadsheets are not, IMHO, an evolutionary masterpiece as they stand.
I think you could make the argument that they’ve certainly evolved to fill a certain niche and that there isn’t a better solution out there in all cases, thus selection pressure hasn’t been able to greatly mold them in recent history (to keep going with the evolution idea), but any argument which claims that spreadsheets are incredibly good at all that they’re used for is problematic.
For all of the power that spreadsheets give people, they’re also problematic. How? Why? Human error and bugs, of course. A quick googling yeilds a lot of interesting reading:
from:
http://www.theregister.co.uk/2005/04/22/managing_spreadsheet_fraud/
“So how much money is the Fortune 500 wasting annually? It is a simple sum:” … ” [t]hat amounts to just shy of three quarters of a billion dollars. And is that anywhere near realistic? No. It is probably safe to say that corporate America, for example, loses in excess of $10bn annually through the misuse and abuse of spreadsheets. That’s a big number: it suggests a problem worth managing.”
Also, see:
http://panko.shidler.hawaii.edu/SSR/index.htm for a whole discussion of issues from errors to compliance challenges.
A choice quote: “Not long ago, recommended practice in medicine involved the use of leeches. This and other dubious practices continued for generations because there was no science to shed light on ineffective (and often harmful) recommended practices. I hate to say it, but I think that recommended spreadsheet practices are at the same stage of development today. ” ( http://panko.shidler.hawaii.edu/SSR/Mypapers/Rant-on-science-in-SS-BPs.pdf )
– Sometimes stuff sticks around for a while because we don’t know any better… (e.g.: asbestos, lead paint, the Chicago Cubs, to name a few)
Spreadsheets are as prone to errors or more prone than regular programming tasks and yet we’ve got all sorts of tools for coding that just aren’t available for spreadsheets (i.e. testing strategies/tools, IDEs (you can make arguments about this point, I guess), easy revision control (try getting a simple diff of a spreadsheet represented in text form that makes clear sense), the list goes on…)
Now, as a simple data management tool or as a way to do a quick one-off sort of thing, spreadsheets are decent, but to do anything complex, their value rapidly decreases and the liability they pose increases.
Okay, now I’ll go read everything past your second paragraph, because I’m sure that your actual topic is really cool and that you didn’t mean for anyone like me to go off like this on spreadsheets =)
Thanks for the comments?
hehe.
Yes, totally agreed, spreadsheets aren’t actually the greatest thing ever, but for one-off “let’s look at this data” tasks, they’re great.
nice article, but described way is very primitive. CSV is good to export some data and import them. It`s great to exchange data between applications. not to display them. usually “row” export export is first request from users. secondly we have requirement “I would like export formated data, with header, merged cells, different fonts”. and it well cause full rewriting. It is better return usual html page with table in it. set content type and excel will open it.
Liked the article, hope to see more.
@cencio,
I didn’t realize that…good tip, thanks. Any idea if it works with earlier versions of Excel? Or is it only the latest and greatest? Regardless, great tip. Thanks.