[erlang-questions] Erlang is the best choice for building commercial application servers

[Edmond Begumisa]

On Tue, 13 Mar 2012 03:10:00 +1100, Miles Fidelman  
<mfidelman@REDACTED> wrote:

> Edmond Begumisa wrote:
>> My impression is that it boils down to marketing.
>>
>> .Net gained acceptance in the business world fairly quickly by *both*  
>> developers and managers.
>>
>> If Erricson were in the business of selling software (which they are  
>> not), and put together a group of spin doctors with a fat budget,  
>> things would be different.
>
> The important part is "in the business of selling software."  The  
> critical business consideration, when selecting infrastructure, is the  
> level of continued support you can expect.  IBM is going to be around  
> for a while, so is Microsoft.

I used to believe that but not anymore. I'll give example:

There are many businesses that invested heavily in the previous iteration  
of MS development infrastructure (COM-driven Visual Studio 6 and related  
tools), and then suddenly had the rug pulled from underneath them in 2002  
when .Net appeared and they were expected to rewrite/migrate much of their  
code (I worked for such a victim, and gathered many now-worthless skills).

It became clear to me that choosing development infrastructure because the  
vendor will "be around" guarantees you zilch continued support. Your  
business is at the mercy of their business decisions. Yet strangely, I  
watched as many businesses with years of complex COM-driven code, grumbled  
for a while, then eventually engaged in expensive time-consuming  
migrations to .Net or started new projects in .Net!

"Why would a business do this?", I pondered, "this vendor has just  
illustrated that they can drop you without thinking twice! Why not just go  
to someone else entirely? And why would a programmer learn C# which may  
well be worthless at some random point in the near future?"

The only sane answer I could come up with was successful marketing and PR.

Microsoft succeeded in skillfully convincing people in this sector that  
re-writing their code in this new environment would allow their software  
to fly to the moon (managed code, reduced TCO and all that hoopla),  
despite having just illustrated that they could well decide to drop .Net  
anytime and move to something entirely different. Suddenly there were lots  
of .Net projects, .Net jobs and lots of .Net developers after those jobs.

I concluded that marketing and PR affect this sector a lot more than  
people realise. People think they are making decisions based on  
engineering and business sense like "continued support" but there's a lot  
of evidence that this is not the case. Many of the "decisions" are being  
fed to them.

> If you prefer vendor independence, you can be pretty confident that C is  
> going to be supported by someone, so is Apache.
>

Yes, I've come to the realisation that open and open-source development  
tools/infrastructure are the only ones that can guarantee you continuity,  
even if your business has to provide that continuity itself.  
Counter-intuitively, big commercial vendor names are actually riskier.

> My sense is that the Erlang/OTP ecosystem has become pretty strong in  
> recent years - enough so that, if Ericsson bailed out, Erlang would  
> continue.  A few years back, that might not have been quite as good a  
> gamble.  (Comments?)

I agree. I'm not big on "industry standards" (i.e. popularity) these days.  
But many businesses are, and not entirely for the reasons they think they  
are. Getting those on board with Erlang I think is largely a matter of  
playing the game the way other vendors play it -- lots of marketing, lots  
of PR, lots of spin all requiring lots of money. But Ericsson's in an  
entirely different game.

- Edmond -

>
>
>


-- 
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/