Friday, 29 October 2010

XML for dummies

Is there an "XML for dummies"? I must get a copy for our favourite telco.

Once again, it seems, they have a simple text substitution in an XML message they pass on where they add our company name and forget to escape the &

How does a big company make such basic errors?

14 comments:

  1. http://www.amazon.co.uk/XML-Dummies-Lucinda-Dykes/dp/0764588451

    ReplyDelete
  2. > How does a big company make such basic errors?
    Much more easily than a small company, but I'm sure you knew that.

    ReplyDelete
  3. Rename your company from "Andrews & Arnold" to "A&A');Update billing set amount = amount/2" and see if the are as good at filtering SQL as they are at generating XML :)

    ReplyDelete
  4. You should change the company name to, literally, 'Andrews & Arnold Ltd'. That should confuse a few people!

    (That made me curious to know what characters are allowed in a company name. Could you be really awkward and incorporate ερος Ltd? It turns out that you can't. You are limited to Latin alphanumerics and a few punctuation marks.)

    ReplyDelete
  5. I know that you're allowed superscript numerals, although most of their systems then struggle with it and "de-superscripts" it (which causes no end of confusion, of course...)

    ReplyDelete
  6. That's funny, it seems Google have a similar problem to BT. I was trying to say, 'Change the company name to, literally, "Andrews ampersand-amp-semicolon Arnold Ltd".'

    ReplyDelete
  7. Big telcos probably make these basic errors in the same way that small ISPs do...

    You might want to look at the mark-up of some of the aaisp sites, before nipping out for a copy of 'Humility for Dummies'.

    ReplyDelete
  8. Very funny. That's actually HTML not XML for a start.

    ReplyDelete
  9. I wouldn't flatter it with 'HTML'. I think 'tag soup' is closer to the mark. XHTML is supposed to be XML anyway, so it's too late to say 'it wasn't XML' after you've stuck an XHTML doctype at the top.

    I haven't found an A&A page which does validate yet, so it doesn't matter where you start. ;-)
    status.aaisp.net.uk is fairly spectacularly malformed, although of course nothing to touch clueless.aaisp.net.uk/san.cgi, which I can't have complained about for at least 18 months.

    ReplyDelete
  10. I once had to explain to a big supplier (begins with a T) that parsing XML using regex and expecting the XML requests I send to them to be 'pretty printed' was a bit wrong.

    @Will - there's a big difference between poorly formed API XML and poorly formed content X/HTML. Especially when being able to perform a line of business function where it's pretty much understood that the XML must be parsable.

    ReplyDelete
  11. @kevinkenny - my point was merely that simple mistakes of carelessness are easy to make, and
    that insulting other people who make the same kind of mistakes as oneself just creates a hostage to fortune.

    The technical details are irrelevant, though I am unimpressed by the counter-arguments so far...

    ReplyDelete
  12. To be honest I have not really made any counter arguments. I said "fair enough". Our web site has errors in it.

    As Kevin said there is a huge difference between a poor web site, and an invalid XML document generated internally by an organisation and sent to another part of that organisation that then rejects it due to the errors the first part introduced, and being seeming unable to actually fix their error. That is just inept whereas, I think, we have just been a bit sloppy.

    I'm happy to take criticism of our web site and work on it though. So thanks.

    ReplyDelete
  13. By the way, I checked http://aaisp.net.uk/ main page as I thought it validated. Found 2 errors, and in my defence they were in the TBB award image and link which was copy/pasted from their email and have now been corrected.

    I am sure other pages do not validate and have various problems though.

    ReplyDelete