As a general rule, when I program in J (or really in any language), I
tend to program in terms of what I want to happen to the data. I
expect the data transforms to be understandable, and I expect that any
good programmer could rewrite the code once they understood the data
and how it gets changed.

In general, I'm a fan of that style of programming.

Thanks,

-- 
Raul


On Sat, Sep 26, 2015 at 12:41 AM, Jon Hough <[email protected]> wrote:
> Well, one point they make is the awfulness of shared mutable state between 
> threads. I suppose J solved that by being single threaded.
> Others...
>
> Bad architectural designs and abstractions...   well, I don't know what kind 
> of abstractions are generally used in J. OOP patterns seem to be used little 
> (OOP itself, seems to be used little). J seems to abstract everything in 
> another direction, by abstracting algorithms and then letting them be 
> composed in different ways and on different datatypes.
> Difficult to understand solutions (over engineering)...    From what I see 
> (bearing in mind I only use J as a hobby), there is little overall structure 
> to J programs, in a Design Pattern sense. That is actually one reason I like 
> using J, I can just get straight to the solution, with no ceremony, cruft, 
> taking care of incidental issues... but then again, it is difficult to argue 
> that a super long tacit verb is easy to understand or extend or modify.
> Bad use of agile and buzzword methodologies...    I don't know if "enterprise 
> J" users even use these kinds of methodologies. I can't see it being too 
> different from other languages in this regard though. A standup meeting is a 
> standup meeting after all. Incremental changes and feedback cycles don't 
> change much with language, I suppose.
>
>
>> Date: Sat, 26 Sep 2015 00:08:44 -0400
>> From: [email protected]
>> To: [email protected]
>> Subject: Re: [Jchat] Interesting talk "How did we end up Here?"
>>
>> I'm only 17 minutes into it but they seem to be asking a lot of questions
>> and posing problems to which the array-language community has answers.
>>
>> On Fri, Sep 25, 2015 at 11:39 PM, Jon Hough <[email protected]> wrote:
>>
>> > I thought this youtube talk from the Goto conference might interest some
>> > people here
>> >
>> > https://www.youtube.com/watch?v=oxjT7veKi9c
>> >
>> >
>> > Essentially, the two speakers are musing on why everything in software
>> > development is so terrible, convoluted, messy etc.
>> >
>> > It's quite long, but might be of interest to some people.
>> >
>> > I enjoyed the quip "The internet is basically in debug mode" as we are all
>> > passing around text data (JSON or XML etc), since I've been looking into
>> > protobufs (not with J!) binary serialization of data.
>> > ----------------------------------------------------------------------
>> > For information about J forums see http://www.jsoftware.com/forums.htm
>> >
>>
>>
>>
>> --
>> Devon McCormick, CFA
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to