The other week we were connected to a remote server trying to debug some ASPX code. That’s remote as in: on another continent, and not using the fastest connection. (It was designed to be fast for a remote population of users, not us.)

A good way to debug the code without switching on debug errors for everybody is to RDP onto the box and browse it from there.

This didn’t work because, despite Windows 2008 IE ESC (Internet Explorer Enhanced Security Configuration) not being enabled, it was blocking cookies, and every option I could find to try and turn it back on was either disabled, or did nothing.

‘Cos you know, browsing localhost (which has your own code on it ferchrissake) is a terrible security risk. Sigh.

In a rush to resolve it, we eventually resorted to downloading and installing Firefox on the machine. (I love Chrome, but its automatic updates and things scare me slightly; probably not good for a server. Some day I’ll dig around and figure out if it can be turned off. Come to think of it I should make sure FF doesn’t do the same thing.)

Firefox brought its own problems. Straight after installation, it decided to load up some hideously slow (on this less than ideal link) page with video. No, actually, two pages — using two tabs — and despite it proclaiming how fast and responsive it is, wouldn’t respond. Blargh. Thanks a bundle, Mozilla.

It must have taken a minute or more to come back — a long frustrating time when you’re in a hurry.

This solution did, however, work — we could finally see the debug messages, thank goodness.

Is there an official way of getting IE to behave itself?

Two programmers looked at each other, thought about how they value their time and the response was a no-brainer:

“We’re programmers,” I explained, “we’ll take the paper form.”

“There’s a phone number you can call if you have any trouble filling out the electronic form” reassures the collector.

Cathy thinks: “Sure, that line won’t have any trouble when twenty million Australians simultaneously log into the web site to fill in the forms via a broken SSL link, using IE specific controls (that only work under some versions Windows assuming they’re correctly patched and have the right libraries loaded), demanding full round-trips to the underspec’d Windows servers to populate unnecessarily complex custom controls, some of which will no doubt demand Flash or COM. Come to think of it, it probably won’t even be web based, and we’ve only got two Windows boxes, one of which is tucked under a table (Yay! Census night on the floor swearing at the ABS’s programmers!) and the other has a screen resolution that went out with buggy whips (I’ve had programs barf and refuse to run because the resolution was unacceptable).”

We chose paper. For another view of the world, I’m looking forward hearing to how census night worked for Daniel…

Fifteen years ago I did a Microsoft certification thingy, and now they want me to do a satisfaction survey on it – for no compensation.  I think not.  But I notice an unsubscribe link at the bottom of the email, so I follow it: http://www.mailingsvcs.com/optout.aspx?type=email&optout=1&service=1&networkid=9001&id=josh@example.com&pid=p53457652, see the Submit button, click on it… and nothing happens.  And then I realise – it needs JavaScript to press.  A button, one of those things right at the heart of HTML 2.0.  What is this, amateur hour?  Turns out, yes it is because if you follow the hacked URL above — which if filled with bogus data — and click on the Submit data, the back end proceeds happily without validating any of the data, and asks you another question before confirming that it’s done:

We’re sorry you no longer want to receive e-mails from us. Please allow one week for us to process this request, during which time you may still receive e-mails from us. We apologize for any inconvenience.

To help us improve our service, please tell us the primary reason why you no longer wish to receive our messages:

There appears to be some kind of problem with their computers.  Last time I checked, the time it takes a computer to remove a record from a database is in the vicinity of “I’m already finished”, not one week.

I’m of the opinion that people who construct software ought to be required to put their name on it in a visible way, so they can go on my list of people to smack in the face when I meet them.  It’s for the best.

class BookForm(forms.Form):
    title = forms.CharField()

def BookView(request):
    form = BookForm()
    return render_to_response('book.html', {'form': form})

With boot.html containing (amongst other things):

<form action="" method="get">
{{ form.as_table }}
<input type="submit" value="Search" />
</form>

Which is great! MVC, separation of data, presentation and business logic. Now, how do you get a CSS class onto that title field? CSS, being the way of separating out the presentation part of a HTML page from the data that’s embedded in it? As above, but chuck it in as such:

class BookForm(forms.Form):
    title = forms.CharField(
        widget=forms.TextInput(attrs={'class':'title-field'}))

Seeing this crunched the gearbox in my mind. All that messy designer stuff, where they make things look nice, that’s worming it’s way into my business logic? Perhaps it’s not so wrong, as the business logic does indeed know that this is a title-field. But it doesn’t quite sit right with me. I’m not convinced it’s wrong, but if you were, you could instead do this in your CSS and HTML:

<style>
.title-field input {background:#ccC68f;}
</style>
<form action="" method="get">
<table>
<tr><td class="title-field"> {{ form.title }} </td></tr>
</table>
<input type="submit" value="Search" />
</form>

Which pretty much forces you to individually place fields — you get to specify the order of fields plus their individual CSS classes.

I’m not sure what the answer is here. Anyone care to enlighten this noob? Bear in mind that there’s a thing to magically tie a model to a form meaning you don’t even need to specify the fields in both the form and model, which you can’t use if you start tossing styles into each field.

Does it really need each available CPU cycle to wait for the other end to do something? Apparently so, in the way it’s coded.

Google: not as clever as the press release makes out.