My tech talk on Tornado (video and slides)(bret.appspot.com)
bret.appspot.com
My tech talk on Tornado (video and slides)
http://bret.appspot.com/entry/tornado-tech-talk
10 comments
Every time i see some code examples of tornado i am getting even less exited about it.
Nothing new there, nothing neat, it just doesn't look as modular and flexible and object oriented as other frameworks.
Am i the only one?
In which case, write your own :) It's just code.
NO! Don't write your own!! That's what got us into this mess in the first place!
Find one you like, and do what you can to support it in some way.
Find one you like, and do what you can to support it in some way.
Why? If I'm designing a new table I want to make out of wood, should I go find someone elses designs, ready made legs etc and use them?
I'll write my own if I want to thanks.
The 'mess' is relying so much on 3rd party libs in the first place, without completely understanding them.
Write something, and learn massive amounts in the process.
I'll write my own if I want to thanks.
The 'mess' is relying so much on 3rd party libs in the first place, without completely understanding them.
Write something, and learn massive amounts in the process.
If there is honestly nothing that does what you need, fine.
I'm just saying the NIH-syndrome is a big problem for the FOSS community, and leads to the fracturing of attention. Far too many talk a big talk about how collaboration is the key to FOSS ('a million eyeballs', etc), but then turn around and start their own project in solitude. Sure you can learn massive amounts -- but how much of that time is spent making the same mistakes someone else already made?
I'm comfortable with telling a programmer "NO! Don't write your own," because I know if s/he really wants to, nothing I say will change that. The value of getting them to stop for a minute to think if it's necessary is worth it.
Any competent programmer can write their own X. But it takes a master to help someone else's X succeed.
I'm just saying the NIH-syndrome is a big problem for the FOSS community, and leads to the fracturing of attention. Far too many talk a big talk about how collaboration is the key to FOSS ('a million eyeballs', etc), but then turn around and start their own project in solitude. Sure you can learn massive amounts -- but how much of that time is spent making the same mistakes someone else already made?
I'm comfortable with telling a programmer "NO! Don't write your own," because I know if s/he really wants to, nothing I say will change that. The value of getting them to stop for a minute to think if it's necessary is worth it.
Any competent programmer can write their own X. But it takes a master to help someone else's X succeed.
>> "But it takes a master to help someone else's X succeed."
I think it often takes more of a politician. Trying to satisfy hundreds of different users and use cases is often orders of magnitude more work than just writing something for your particular use case. Not to mention all the ideology/style/language wars/etc etc
I think it often takes more of a politician. Trying to satisfy hundreds of different users and use cases is often orders of magnitude more work than just writing something for your particular use case. Not to mention all the ideology/style/language wars/etc etc
Not everyone agrees that this is a mess.
Which is why we should rally behind the original async framework, Medusa: http://www.nightmare.com/medusa/ ;)
And if you don't like it...?