[pypy-dev] pypy on Tim Bray's blog

[Martijn Faassen]

Hey,

Jacob Hallén wrote:
> Den Tuesday 13 November 2007 19.42.03 skrev Martijn Faassen:
>> I'm convinced the body count of Python users is probably still quite a
>> bit larger than Ruby users. I also have a strong suspicion far more
>> people are using Python web frameworks than there are people using
>> Rails, hype aside, but I may be wrong about that.
> 
> It doesn't really matter how many users you have, when someone measures
> vibrant by how many excited people he meets at a conference. There
> were 600 at a recent Rails conference, while Pycon gathers 400.
> 
> This is the way Tim Bray sees things and it is not our mission to tell
> him that he is wrong. It is our mission to get him to pay for PyPy
> development. He is the customer, and the customer is always right, ;-)

He's not *my* customer, so I can say whatever I like. :) Since I'm not 
your customer either, obviously I'm *not* always right, which means we 
can actually have a conversation.

>> I think making it practical and useful to run a PyPy-based Python
>> interpreter in production projects is probably the best shot this
>> project has at becoming more self-sustaining. You're close.
>>
>> So I'll definitely complain if you spend a lot of time on Ruby (or
>> Smalltalk for that matter) before Python's all the way there. I think
>> that'd be a bad idea for a whole range of reasons - get the last 10%
>> done for Python first before you take even more on your plates. We all
>> know that last 10% tends to be the hardest part. Focus is important. If
>> you can't make that work for Python, you'd have a hard time making it
>> work for any other language too (or convincing people that you can). Of
>> course I realize I have no real voice in this project, but that's my input.
> 
> The reason we are not there yet for Python is financial, not due to a lack
> of focus. (And what yo may have seen as a lack of focus during the funded
> period was the totally opposite. We focused on delivering what we had
> promised the EU, not on what would be the fastest path to be a viable
> alternative to CPython.)

> If we get funded to do Ruby, this will benefit Python, but the focus will be
> on delivering what the one who pays wants out of the project, not on
> making a better Python.

I'm not the customer, but you'd have an awfully hard time selling this 
story to me. If I were looking for a technology to make Ruby work 
faster, and I got some people saying PyPy, but um, we don't have an 
actual completed language on it, not even Python, I would be skeptical.

Moreover, if you seriously start developing a Ruby interpreter before 
Python is done, I'll call lack of focus. Similarly if you seriously 
start developer a Smalltalk interpreter.

The PyPy project is researchy. Its people likes to play around with 
research topics, exploring possibilities, and of course that's more fun 
than to do the grind of making stuff work in production. (I'm not 
implying you haven't gone through a lot of grind; I'm just explaining 
the psychology that increases the risk)

I don't want to talk about the history here: whether there may or may 
not have been a lack of focus in the past, during, say, the funding 
period. I understand the requirements of this period. But this project 
is, in my opinion, at present at *risk* of suffering of a lack of focus.

I think this focus should be a production-ready Python interpreter. 
That's what you set out to do. It's not done yet. If you seriously start 
building other interpreters (beyond, perhaps, some exploration), you'll 
get sidetracked from that indefinitely.

So, if you can get a Ruby interpreter funded at this stage, fine. I'm 
happy for you. But I will give up hope I'll be able to use any 
PyPy-based interpreter for the foreseeable future. Python, Ruby, or 
otherwise. (and you'd be free to prove me wrong, of course)

Regards,

Martijn