Good to see some of the concerns addressed.
"
Now so far I do not see companies really putting effort so may be
nothing will happen but this will not be
because of us. :)"
This is the chicken and egg situation. Bar highly motivated startups with
some money in their pockets to splurge on with. The average co consists of
average managers who want no risk..!. They want a technology they can blame
for its shortcomings/ the support offered by another co if they are stuck
for a fix. But in most ( I would say 90+% of timeline) cases the business
continuity should not be affected.
The average company will probably not invest their time on a technology if
it does not meet the bar set by the current technology.
Let me take Spring Architecture as an example in the Java world. J2EE was (
and to an extent is) entrenched in the world of Java enterprise. Way back
about 8 yrs back or so .. Rod Johnson started his foray in to simplifying
the complexity of J2EE with his framework. I would say through atleast 4+
yrs of the 8 he would have close to nil support from any company and like
the Jim Collins "Good to Great" simile built up the giant wheel momentum now
to engage nearly all known companies to use Spring all through instead of
J2EE except in the niche cases. Its is an instruction to notice how Spring
got interfaces to nearly all of Java connected that would be possibly needed
for a medium enterprise case and then went into the depths/ specialization
etc.. that is breadth first and then the depth.
So I would say "WE" (including myself as a avowed Smaltalker) need to keep
trying and pushing for a concerted go at getting Pharo up there.. and
possibly the "GiantWheel momentum" will kick in with first a few co's and
then more.. to push this rolling with god speed to its eventual
greatness..!!..
And that indeed is happening and its suprised me how far Pharo has already
rolled and is building a momentum that is sure to go far if I can put my
little effort as all others to get some of the minimal frameworks
integrated.
We have either of two approaches to take: meet up to the current bar set by
Java/ .Net world in terms of programming baseline ( as I listed in the prior
mail) or take a radical approach that differs so much and offers so much to
pull in others..like Rails did. I would say if we are interested in the
numbers game I would choose the former, if we wish to retain the
intellectual high ground and move on the latter is fine..
To get the numbers to have an interest in Pharo I will go back to my
charter for Smalltalk spread in Universities / Colleges ( the underlying
reason I started SmalltalkIndia) and see how far it can be resuscitated to
create a mass base of users ( even if they are amateurs) and then hope a
good percentage of them retain a greater interest to contribute spare time
to improve the frameworks in Pharo.
*************************
Just count how many smalltalkers we can get in a low cost centers who can
code.. well
Contrast this with how many Java programmers you can get.. can manage with
google/ info base available
Count the external frameworks open source developed , tested and trustable
to be used in production code from Java nearly all free.
Count the same for Smalltalk
App servers.. comparable to Websphere/ weblogic/ Tomcat / lots of others,
not to mention messaging queue, transaction control , JDBC like framework
for nearly all DBs with high performance guaranteed, the list goes on..
The support logistics in terms CMS: viz SVN kinds, better integration /
build systems like maven etc.. and evolutions in terms of frameworks that
Java has spewed.. .Net in its Visual Studio et als..
Good brains together can counter all of the above arguments, but that is a
limitation by itself, you cannot get good 25-50brains in one premises to
work together on one single product, even if you do have them you cannot
easily replace them with new recruits and be cost effective in general.
From an ease of development and risk free managment angle, I find this an
impossible proposition to convince any mgmt to take up Smalltalk for their
dev.
The target is the average developer, the risk averse corporate entity in
all its evolution whether its .Net or Java.
For all the reasons above, corporate use of ST is a difficult game for
niche languages like Smalltalk, but a target I would like to see achieved in
the near term..
*********************************
-Skrish
On Sun, Aug 29, 2010 at 8:32 PM, Sudhakar Krishnamachari <
[email protected]> wrote:
> My two cents long time in my blog on exactly the same subject:
>
> -Skrish
>
>
>
_______________________________________________
Pharo-project mailing list
[email protected]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
