[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Y Store now C++

DLWeinreb@attbi.com wrote:

> Michael Vanier wrote:
>> Five years from now, when people who took scheme in CS 1 have 
>> graduated and
>> are working in industry, some of them will stumble across problems in 
>> java
>> or C++ and say to themselves "this was so easy in scheme; we'd just use a
>> closure".  Then they'll start playing around with languages that support
>> closures in their free time, maybe choosing one of them as a scripting
>> language to use with their work, and suddenly industry is using more
>> advanced languages.  Of course, the key to the success of this is to not
>> tell your boss what you're doing ;-)
> Well, if they start checking in product code written in some language 
> that isn't the language
> that everyone else uses, they'd better have a pretty tolerant boss.  I 
> have never worked in
> any company in which it was acceptable for one programmer to choose a 
> language of his
> or her choice to write product code in, without getting buy-in from 
> relevant responsible
> parties, and that includes Symbolics.
Wasn't it this list where someone suggested the "rapid prototyping" way 
to get alternative languages accepted?  The idea was that you do an 
initial prototype or proof of concept, justified to your boss as a quick 
mock-up to flesh out the ideas and validate the requirements. 
 Presumably you do that extremely quickly because you're using good 
tools and you're a way-above-average programmer.  Or maybe they're 
way-above-average tools.  :-)  You show it to your boss, get agreement 
on the features, etc.  Then you point out that what you have is actually 
a working implementation, not just a mock-up.  And that you could spend 
time rewriting it in the company's favorite accepted language, or you 
could spend time adding features, tuning, and otherwise polishing your 
weird-language implementation.  Now you might get the boss to approve 
checking the code in, for schedule/time-to-market reasons.

Of course the Yahoo store experience suggests that reimplementation will 
eventually happen, but by then you'll have moved on to another project.  :-)