WTF Company
Written by Jevgeni Kabanov on March 23, 2008 – 9:41 pmNext day I asked my mentor if I could replace the regexp based XML-parsing with some standard classes to make it cleaner and easier to extend. He said no, because their parsing works just fine, thank you. — Somehow this (part I, part II) is even scarier that reading it on DailyWTF. Does anyone recognize your own company? I sure hope not!
Posted in humour | 6 Comments »
6 Comments to “WTF Company”
Leave a Comment
Additional comments powered by BackType
March 24th, 2008 at 5:51 am
You’re telling me that the company you’re working for is using RegEx to extract data from XML? They’re not even using XPath or something more logical?
Isn’t it time you quit your job and moved on?
Best Regards,
Richard L. Burton III
March 24th, 2008 at 7:00 am
Thanks God wasn’t my company, check out the links. Just quoting.
March 24th, 2008 at 12:55 pm
Let me get this straight: your company has working solution that uses suboptimal technical software stack. It works 99% of cases and does not lead to lost/corrupted state.
Now you offer same solution, in business sense, that uses perfect techology stack, and it costs X amount of money (time=money) to complete.
You should ask yourself what would you do if that company was entirely yours.
Obvious troll is obvious
March 24th, 2008 at 2:37 pm
It’s always juggling short-term benefits versus long-term benefits. Some of the choices are just widely known to be bad in the long term, because so many have run into them. As an owner I need to avoid such decisions unless I know that long-term isn’t relevant (e.g. I’m selling the company off in half a year).
March 25th, 2008 at 12:03 am
Well that depends on context. For mission critical parts of the system, such as lets say transaction management, bank account services, of course you got to use all possible best practices, however, for XML parsing, I fail to see business benefits, unless of course XML schema upon which XML parsing is based changed very regularly, which would requires modifications to that unfortunate in technical sense, but good in business way of doing things.
March 25th, 2008 at 12:36 am
But if you read the posts I’m pointing to it’s clear they had to modify it a lot to absorb the changes.
The problem is that regexp is an inherently bad way to parse XML, which is not a regular language. Therefore you can almost guarantee that every small change (and there are always changes) will cost a lot. Whereas with a standard parser it would cost nothing to begin with.