You'd think you'd think

[j4]

So when I find an incredibly incorrect (i.e. containing at least 3 errors, depending on how you count them) URL in this year's Awards issue of the Reporter, I email the Reporter editrix and say "I say, this is a bit wrong," and she emails back cc:ing in the person who should have Got It Right (let's call him Mr Wrong) and saying "Yes, this is wrong. Mr Wrong, you should have checked it." All well and good, though I suspect Mr Wrong wasn't too pleased to be outed to me as the culprit.

But then I decide "What the heck, I'll open this interesting-looking can marked 'Worms' check the rest of the URLs as well, just in case." And I find an alarming number of other errors, and embarrassingly many of them are Cambridge University department/college websites. Which is a good thing in some ways, because it means we stand a chance of telling the people who supplied the wrong URLs, and in some cases getting them to put redirects in (this is an annoyingly high-profile place in which to publish incorrect URLs, and we could do without a gazillion more "help pls wear cna i find this page IT ARE NOT WORK AND V URGENT NOW kthxbye" emails than we already get). And yet it's a bad thing in other ways, not least because it gives the impression that any attempt to access web-based information at this illustrious institution will return a big fat "404: arse not found with both hands" error.

And then I check one obviously-incorrect URL, and discover that we've already put a redirect in from a slightly-differently-wrong version of the same wrong URL, because IT APPEARED INCORRECTLY LAST YEAR, and you'd think (you'd think you'd think you'd think) you'd cocking well make sure you GOT IT RIGHT after that, but no, but no, but no.

Help, pls. Wear cna i find willpower? IT ARE NOT WORK.

Comments

[caramel-betty.livejournal.com]

I occasionally have similar fun at work, when the marketeers put out any old random URL they haven't told us about, and don't tell us they want it to work, and the first we get is something like "Can you put a PDF of this advert on the website please?" and if we read the advert we spot the URL on it and then have to scramble to find out if this advert has already been sent to customers/websites.

Mutter.

[anat0010.livejournal.com]

Imagine how Mr Marketeer sees this. He puts random URL in ad, publishes ad and hey presto the random URL works. Incredible he thinks, I'll try this trick again ...

[j4.livejournal.com]

Yes. This is why, even if politics eventually dictate that we have to make the URL exist to fit the paper publication, we still kick up a big fuss about it. Every time.

Maybe one day it will sink in...

[caramel-betty.livejournal.com]

Not even close. They get their ear bent heavily from all sorts of different people every time they do something like that, and they still do it and don't give us any advance warning.

[fanf]

They do this with email addresses too.

[lusercop.livejournal.com]

My willpower is at an all-time low at the moment, I have to admit - today's fun was to delete a day's work just as I was about to commit it to our source control: "hmmm, that shouldn't be there - remove it", "copy file from source loca... bugger". I feel this will about set the tone for this week :-(

[jiggery-pokery.livejournal.com]

*giggles*

If it's any consolation, do know that you are on the side of the angels.

[besskeloid.livejournal.com]

"404: arse not found with both hands"

*cackle*

[pseudomonas.livejournal.com]

I might just have to steal that line :)

[arnhem.livejournal.com]

In theory, content manglement systems with a suitable model of workflow (separation of editing, automated checking, approval and publishing) are the solution to all such ills.

In practice, they require more effort in analysing existing practice, in order to implement them in a way that doesn't blow up in everyone's faces, than most departments are prepared to / can afford to invest.

My impression is that it's not that problematic, technically, since there exist packages that do what is wanted pretty well out-of-the-box; it's the understanding of "business processes" (gah, hate the phrase) that's the awful bit. In particular when one of the causes of resistance by organisations to such analysis is the fact that it'll show up how awfully things are done.

[oldbloke.livejournal.com]

The minimal version where any TDoH can write pages but a dedicated team who can check them properly are the only ones who can copy them to the live site would be better than the mess J seems to be stuck with, though.

[j4.livejournal.com]

That is what we've got, more or less. The problem is that the Reporter is published in paper format as well, and we can't change the paper version. We also can't necessarily work out what the correct URL was meant to be, because the information comes down a chain of about a million people and we don't always have any way of getting back to them. So we tell the person at our end of the chain that something's wrong, and it never gets back to the originator.

The other problem is that the "dedicated team" at the moment consists of, um, me.

We're trying to devolve responsibility for things like actually KNOWING YOUR OWN ADDRESS to the people who supply the information, so when they say "It's wrong on the web!" we can say "That's because you sent us something wrong to put on the web!" (but still fix it, obv) but it's a slow battle. :-}

[oldbloke.livejournal.com]

Ah, yes, paper docs, produced by somebody who doesn't talk to anybody else...

[j4.livejournal.com]

It would be really useful if I could convince people that "we've made this long paper document -- now put it on the web" is all too often a bit like "we've made some jam, please build us a wall with it". Only without the "please", usually.

[oldbloke.livejournal.com]

Walls: bricks held together with mortar, or mortar spaced out with bricks?

[arnhem.livejournal.com]

The problem will be trying to convince someone that the paper format ought to be derived from the online one (or both from a common, managed, source).

They won't understand.

<earworm>dedication, that's what you need ...</earworm>

[j4.livejournal.com]

We tried that. They understand the concept, but (for a variety of reasons, some slightly more sensible than others) they're not willing to change their workflow in any way in order to achieve it.

There's also the fact that the paper version has been in existence since about 1850, & things acquire a kind of weight and Officialness over the centuries. It's sometimes difficult trying to convince people that this crazy new electric stuff is here to stay.

[camellia-uk.livejournal.com]

For freshers fair, our lovely publicity officer printed somewhere around 500 flyers for our society.
And no, she hadn't checked the society's website address was right, since she didn't really understand what it was.

They all had to be changed. By hand.
Somewhere, there's a manager thinking "...so all we have to do is go round to the house of every user who sees this website, and change the URL in their browser manually!! great plan!" :-S

[pseudomonas.livejournal.com]

We had that a while back with a concert we did, where someone along the chain decided that all URLs started with www so we were listed as http://www.users.ox... rather than http://users.ox... Grrr.