[Ironruby-core] How do you convince .Net developers to use IronRuby?
Ivan Porto Carrero
ivan at flanders.co.nz
Mon Nov 2 11:16:32 EST 2009
AFAICT you don't, they'll have to convince themselves. I may have given this
issue some thought before today :)
One of the problems I have with most of the .NET shops these days is that
they are very conservative. I'm having countless discussions on the benefits
of unit testing, stored procs vs OR/M, ...
then the developers have been misled into believing intellisense actually
helps them while all they are missing is a tool that lets them find api docs
fast (msdn isn't it although the new version is a lot faster), because
intellisense gets more in the way than out of it. The same way they have
been misled into believing static typing is the shiznit.
It took me 6 months of daily scavenging the net for an IDE and eventually I
just bought a mac so I could use textmate and be done with it.
Before we can get ironruby to not be perceived as a "hippy" language there
are a few things that need to happen:
1. .net 4.0 needs to be out (preferrably with a service release or 2) so
they can use it from their C# visual studio project.
2. we need a textmate/rubymate like autocomplete in visual studio that will
just tokenize the words in a document or open documents and puts some
heuristics around which one you are most likely to use.
3. IronRuby needs to be labelled 1.0 (preferably with a SP1 stamp)
4. visual studio must not insert a BOM in the files it creates (this is
solved by creating ruby file templates see ironrubymvc project).
5. We need an example application that when compared to the C# one has
obvious benefits. It would be good to have some kind of build off,
unfortunately the human factor makes that non deterministic and subjective.
6. IronRuby startup time needs to get a _lot_ faster, because antonio
cangiani's benchmark might show different results but for me running a spec
suite with ironruby and one with ruby 1.9 is a difference of 2 minutes vs 10
seconds, mostly because of startup time. but that totally kills the rapid
feedback cycle you want.
7. blog, speak, write, convert them one at a time I say :)
I see ironruby creeping in the enterprise via 4 ways:
1. using as a test framework for their apps (and even then perhaps only the
fan boys at first)
2. a much better nant because nobody likes writing xml
3. for rules engines where the user can define the rules, or as plugins for
4. Quicker prototyping
This is one of those things that isn't going to happen overnight, and the
early hype around IronRuby didn't really help the situation. So that means
we still have a lot of work todo. Writing, blogging, speaking about IronRuby
is a necessity but also making sure the experience is optimal makes a big
If you want to be an agent for change you'll have to realize that that is a
very frustrating position to be in because all you will face is resistance
-being in the front line and all-, but if you take a step back then you'll
see that usage of ironruby has grown quite a bit in recent months and as the
implementation progresses it will increase more and faster ( a little bit
like a snow ball).
Also change doesn't happen on the spot. For example I will get frustrated
and stop my contract but when I talk to that company one year later it turns
out they are actually implementing what I had been fighting for the year
before. I lack patience and tact to be really good at changing mindsets :).
There is some ruby envy in the clr community though. I say ruby envy because
they refuse to move out of C# but wish C# has all the things ruby has but
with the wonderful C# syntax.
To give an example of the mindset:
The microsoft belgium people almost laughed at me when I told them I was
going to write a real app in IronRuby when I applied for bizspark.
I don't see command line as a problem because as the guys from sapphire
steel have shown they can integrate that quite good , rubymine and netbeans
show that too. I'm just faster with the command line and I feel like I'm in
control (which in reality I'm not of course)
I also agree with Nathan IronRuby isn't ready for prime-time mass-adoption
Met vriendelijke groeten - Best regards - Salutations
Ivan Porto Carrero
Author of IronRuby in Action (http://manning.com/carrero)
On Mon, Nov 2, 2009 at 4:06 PM, Shay Friedman <shay.friedman at gmail.com>wrote:
> Hi there,
> In the last month I had 3 sessions about IronRuby, all of them in front of
> .Net audience. I really believe in the IronRuby but I find it very very hard
> to pass that to existing .Net developers.
> I try to show the benefits of using IR - getting things done faster (like
> POCs, internal tools), using REPL, using IR abilities from C#, IR and
> Silverlight (like Gestalt), unit testing, RoR...
> Most of the .Net devs are very conservative and are not willing to get out
> of their familiar development environment even when they see the clear
> benefits of the new technology.
> They feel that using IronRuby will take everything they love from them -
> Visual Studio, Ctrl+F5, the sacred intellisense, etc.
> That's about what happens during a session:
> - No Visual Studio integration: 50% of the audience are willing to leave.
> - No compilation: more 25% have just lost interest.
> - Intensive command line work: more 15% are shutting down.
> That leaves about 10 perecent of the audience that just think of using
> IronRuby, most of them decide not to eventually.
> My question is - how do you suggest to present IronRuby to .Net developers?
> and to the team members - does Microsoft expect that existing .Net devs
> will start using IronRuby?
> Shay Friedman
> Author of IronRuby Unleashed
> Follow me: http://twitter.com/ironshay
> Ironruby-core mailing list
> Ironruby-core at rubyforge.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Ironruby-core