I Work With a Pedant

[NickFitz]

I think you'll find "pedant" isn't the correct word.

[suityou01]

Again, my bad. Not injected as such but rendered server side so no security issues here.

You are sheeting me, right?

[Mich the Tester]

What does your architect say? Is it your call to make that this code is completely self contained?

Is ANY code completely self contained, as in, completely insulated from possible outside interactions?

[suityou01]

Is ANY code completely self contained, as in, completely insulated from possible outside interactions?

See post #22

[Gittins Gal]

What does your architect say? Is it your call to make that this code is completely self contained?

Yes, completely my call. The whole app, frankly is self contained and used for such an arcane process that the chances of anyone ever re-using this stuff is one in a million.

So why bother when the main thing is putting it live to satisfy a legal requirement?

It's a bit like people who use SOA for the sake of it.
Any service layer, regardless of the likelihood of it being used by anything else, is duly stuck in a web service layer and everything serialised/deserialised to/from XML. Never any thought given to the impact on performance

[Gittins Gal]

You are sheeting me, right?

Aren't you familiar with RegisterClientScript in asp.net web forms?

[Mich the Tester]

See post #22

It woz a wetowickal kweschun innit.

[Gittins Gal]

I think you'll find "pedant" isn't the correct word.

Oh he is, he's an awful pedant.

Can't keep his sticky beak out of anything anyone else is doing.

I'm sat there the other day happily typing away , unaware of his presence behind me.

"why are you using server side label controls?" he barked ( he has an annoying habit of coming along and stuffing his face in your screen).

Because I need to conditionally apply a CSS class server side. That's why. If I didn't have to do this I'd use a span instead. (now sod off I'm thinking to myself)

[vetran]

listen, record decision, tell him he can do nothing without authorisation. You let him rewrite code, bad PM!

If he is right you discussed it and the team decided against it because of time constraints, if his predictions don't come true then great.

He has a use because he will spot some doozies but he is also hard work, its a trade off.

The speed thing is classic, the definition of 'good enough' has to be accepted by the users. You just test on your slowest link and record it (media encoder is you friend) , play it back to Stakeholder and get them to say yes because to redevelop will take too long. If its too slow later then you have a get out of jail free card. Total time 30 mins and the pedant can do it.

I am the Pedant on a current project, but the points I raise come true fairly quickly after I raise them (I don't even have to prod).

[original PM]

yes indeed we have these - I often find there are two ways to do things

the fully correct way which takes longer but makes the system more robust and future proof.

or

the way which gives the user the functionality they want

after all who knows what the future holds....

also doing the first one commits you to future apps/solutions and therefore future costs.


hmm as always choices choices!