Re: [tlug] Running without Gnome/KDE/xfce/whatever. (was: Ubuntu 16.04-LTS Japanese Text Input)

[Raymond Wan <rwan.kyoto@example.com>]

On Fri, Apr 29, 2016 at 12:33 AM, Stephen J. Turnbull
<stephen@example.com> wrote:
> Raymond Wan writes:
>
>  > In past company jobs, I have worked with vendors that develop
>  > software for us (i.e., I represented the client).  And I was
>  > appalled at what software development has become.
>
> Man, you've been skimping on your reading then.  Start with The
> Mythical Man-Month and read between the lines to realize just how
> SNAFU-ed IBM development practices were in the 50s and early 60s (and
> they achieved World Domination and Fred Brooks got to be Monday
> Morning Quarterback for Life, so it worked out OK ;-).  Then go on to
> Ed Yourdon's "Decline and Fall of the American Programmer" to find out
> just how bad the U.S. industry sucked in the late 80s/early 90s, and
> then his "Resurrection and Rise of the American Programmer", in which
> Eddie-kun discovers in 1999 or so that the rest of the world sucks
> just as bad, so the U.S. still has a chance.  (Incidently, after
> getting multiple accolades for "world-class software development
> organizations" in D & F, Japan is reduced to a single footnote in R &
> R!  Sad, sad, sad.)
>
> Executive summary: While things are horrible now, they could be worse.
> And frequently have been.  :-|


I've read "Mythical Man-Month", but not any of the others.  Perhaps
I'll look up some of those books -- thanks!

But, I have this feeling that the "next" generation of programmers
(i.e., who have just graduated or are in their 20s and make up many of
the small-to-medium sized vendors) are not learning lessons like this.

I fear that the current CS curriculums have traded such useful content
for more "practical" courses such as iOS and Android programming.  I
guess it's good that they learn skills for app programming, but I
wonder if their foundation is weaker.

And yes, things could have been worse but what bothers me is that they
are passing technical decisions on to the customers/clients under the
guise of "customer's always right" / "customer's the boss".
Ok...sure...so why are we paying you for your IT skills?

Something doesn't look right and it's the lessons mentioned in those
books and something else that I can't seem to grasp.


>  > [*]  Yes, it could be a problem with the companies we were
>  > approaching.  We didn't have a lot of money so we could not
>  > approach large vendors -- only small/medium sized ones.
>
> The large ones are often worse, at least in Japan.  The problem is
> that they don't do very much programming in-house, they farm it out to
> subcontractors, which tend to farm it out to subcontractors, which


I don't know about that.  Honestly, we didn't have the funds to pay
larger vendors to compare.  And I think few will have the resources to
pay two vendors for a project to compare.  In the end, you pick one
based on the quotation they give you...and we all know how misleading
(i.e., the fine print) those quotations can have!


> ....  And since all you've got is management and "systems integrators"
> in the chain, who don't actually do design or coding, they jealously
> guard the communications channels and their level of requirements
> documents, effectively preventing the actual programmers from ever
> learning, let alone serving, the actual client's needs. :-(  And of
> course at the very top end of the scale, they want to sell you a
> "cloud solution", not the software itself.  Which is a completely
> different can of worms, and the only problem that a can of worms can
> solve for you is composting.


*ugh*  Don't get me started on how often I've been hearing about the
"cloud". Most people that use the term don't know what it means and
use it to demonstrate they are up-to-date...


> That's why when I have a friend who needs software written, I
> introduce them to Curt (wow, was that really 10 years ago?) or one of
> the other one/two-man shops with serious OCD I know of.  Of course
> there are large projects that require more than such a shop can do by
> itself in time, but largish companies tend to delude themselves that
> everything they do is big, whereas in fact many of their tasks are
> small enough that a lone independent contractor can do them better,
> faster, *and* cheaper.  You do have to pick the right contractor....


Yes, it would be nice if I had contractors on hand to approach that I
trust.  Getting the wrong contractor is possibly worse than doing it
yourself since you spend so much time checking what they're doing.

I've returned to academia so, for the moment, I don't need to worry
about seeking vendors...


> P.S. I was at a talk today where the lecturer put up a chart showing
> that the *pay* rankings in Japan for several languages was headed by
> Python, Perl, Ruby, C/C++, Java (about a 10% hit vs Python), PHP,
> Javascript, C#, Swift, and Objective-C (by now a 25% hit vs. Python).
> I guess if you know two languages you can add the salaries? ;-)


Reminds me of this article in CACM's blogs:

http://cacm.acm.org/blogs/blog-cacm/176450-python-is-now-the-most-popular-introductory-teaching-language-at-top-us-universities/fulltext

I'm not sure if curriculum reflects the job market or vice-versa.
i.e., what influences what.

It would be interesting if such a study was done in Japan.  i.e., like
the table in the Appendix  Has anyone heard of one?

Ray