The Artima Developer Community
Sponsored Link

Weblogs Forum
Marketing Python - An Idea Whose Time Has Come

117 replies on 8 pages. Most recent reply: Jun 15, 2007 5:29 AM by Ian Ozsvald

Welcome Guest
  Sign In

Go back to the topic listing  Back to Topic List Click to reply to this topic  Reply to this Topic Click to search messages in this forum  Search Forum Click for a threaded view of the topic  Threaded View   
Previous Topic   Next Topic
Flat View: This topic has 117 replies on 8 pages [ « | 1 2 3 4 5 6 7 8 | » ]
Jon-Pierre Gentil

Posts: 1
Nickname: zenethian
Registered: Mar, 2006

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 12:45 AM
Reply to this message Reply
Advertisement
For everyone crying out for Guido to take a stand on "one true framework" it should be considered that it isn't his place to do so.

If you're so gung-ho about making One True Framework, email all of the authors of all of the current popular frameworks, and tell them to drop what they're doing and come together and unite to build a single powerful framework that includes the best features of each.

Fat chance it'll happen simply because each will disagree with the other and it'll all fall apart, but hey, worth a try, right?

Bjørn Stabell

Posts: 1
Nickname: whitebear
Registered: Mar, 2006

A boat without a captain Posted: Mar 4, 2006 4:29 AM
Reply to this message Reply
As software developers we Python users love to create, but what Python needs now is a little more consensus on frameworks and best practices.

Most applications today are web applications, and assuming I'd like to use Python, there is no one obvious web application frameworks to chose from. Nobody wants to invest their reputation and money on a loser, it significantly increases risk. Web frameworks for Python is a fragmented market; nobody has critical mass.

Consensus doesn't happen automatically. In the case of Ruby on Rails the creators are just very talented as promotion and marketing, as well as technology, so they drive it themselves. (TurboGears and Django also do good marketing, but they're fighting against each other.)

Apache Software Foundation helps create "consensus" and put a stamp on approval on many Java projects.

Python has neither core people with marketing interests nor an organization to help bring consensus.

We've been blessed by a benevolent dictator with great technical talent (amongst many others, I'm sure), but with little interest in marketing. There didn't seem to be anybody with real decision making power at home in the python-marketing list [1] (which was the only forum I found remotely linked to Python marketing), and even basic suggestions like Python getting a "marketing friendly" logo has been met with silence. (The old bunch of logos sent a message of amateurness compared to the more polished versions of competitors.)

The new Python website [1] is a step in the right direction, but it's nowhere as good a website as Ruby on Rails: it's way too wordy and it does come off as a bit too much of a hard sell.

We're a boat without a captain with it comes to marketing.

Note: This is not a situation unique to Python; many open source projects suffer the same fate as the main people involved do not have a clue about marketing. Some projects do, and, if they're technically sound as well, win out.

What we need is some way to gain consensus around marketing problems and rally enough troops towards solving them. Only someone, or something, with enough respect and authority can do that.

[1] http://beta.python.org/
[2] http://wingware.com/mailman/listinfo/marketing-python

Jerry F

Posts: 4
Nickname: jefe59
Registered: Mar, 2006

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 8:58 AM
Reply to this message Reply
> For everyone crying out for Guido to take a stand on "one
> true framework" it should be considered that it isn't his
> place to do so.
>

Jon-Pierre Gentil:

It may not be his place to do so, but there are certain things you can't change. (At the risk of insulting Guido by comparing him with a professional athlete...) Just as people expect their local sports hero to be a role model, tech people look for the opinions of their leaders. You can ask people to think for themselves, rather than consider the opinion of leaders, but as you say "fat chance".

Frankly, I am your case study. I was reflecting on how I stumbled across this discussion. Here's how: I'm about to take on the challenge of learning something new. It's a substantial investment of my time. Just as Guido mentioned a few weeks ago - he wanted a framework for a project, and looked for advice on getting started. Same here. I want to learn ONE framework very well.

Now if there are 80 frameworks [1] (maybe more if you include RoR, etc), how do I choose ONE to learn VERY well? I could conduct 80 experiments and "think for myself". Or I can narrow the field substantially by trying to choose one with momentum - after all, no one wants to invest in learning a dead product.

So formulate a short list, and use the experts' opinions to select the one.

My short list a few weeks ago was, in this order: Zope, Django, TurboGears, and Ruby on Rails. So I googled looking for Guido's name and each of the frameworks. I found his blog, and formulated an opinion (that Guido prefers Django) based on what I read. Right or wrong, that's what I got out of it.

My point is that PEOPLE WILL DO THIS. Formulate opinions, repeat what they believe (sometimes in bold), and make decisions based on those beliefs. I am a real-life case-in-point. So whether or not it's Guido's place to do so, people like me (right or wrong) are watching. It's not that I am lazy or won't think for myself. It's that I want to spend my time wisely and learn ONE framework really well. To learn a bunch of frameworks well enough to make an informed comparison (without relying on experts) would be a waste of my time. (Now you could argue that Guido's the wrong expert to be relying on, just like it's wrong to look to Professional Athletes for your morality or gym shoe choice. But it seemed natural to me to look to Guido, so I'm sure that others will, as well.)

It'd be great, yeah, if some unbiased reliable resource compared all frameworks thoroughly, and told me that given my project-type, expertise, and feelings toward product momentum, here's the best framework. But no one is doing that, as far as I can tell. So we turn to the respected leaders. (e.g. if Google backs Python, it's worth looking at. If Guido backed Zope by being employed there, it's worth looking at.) In absence of public statements, people make stuff up: "Guido left Zope, so maybe Zope 2 and 3 aren't all their cracked up to be." I was actually floored to see that in Guido's recent search for a suitable framework for a project, Zope wasn't even on the radar screen! It lost MAJOR points right there - here's an ex-employee, a respected leader, and he's looking elsewhere to technologies that he knows less about - that's a major red flag (like I said, people (like me) make stuff up!)

My conclusion, incidentally, after reading numerous posts is that my short list is now Django and RoR. I eliminated Zope and TurboGears specifically because I "read between the lines" that Guido prefers Django to TurboGears and Zope. All it really took was his Django vs. Cheeta post of 31-Jan [2], and a few references that he didn't want to use XML as a language (which - making stuff up here - are probably reasons why Zope and TurboGears are not good choices)[3].

My point is that in absence of formal statements, people will try to read between the lines, and make decisions on it. Guido prefers Django. There - see I just made it up. That's the conclusion I drew after researching it. So momentum is in the Django camp by an industry leader. That's the tipping point. I will now compare Django with RoR for use on this project. Django will most likely win, because I know and love Python, and feel that learning the framework will be easier without having to learn the language too.

For the benefit of everyone else doing Google searches like I did, maybe they'll locate this thread if I mention once again the opinion / conclusion that I arrived at: Guido prefers Django. And if you read in on the internet, it must be true, right?

[1] http://www.artima.com/forums/flat.jsp?forum=106&thread=146606&start=15&msRange=15
see Guido's post timestamped Jan 31, 2006 6:01 PM
[2] http://www.artima.com/weblogs/viewpost.jsp?thread=146606
[3] http://www.artima.com/weblogs/viewpost.jsp?thread=146647

Andre Roberge

Posts: 6
Nickname: aroberge
Registered: Dec, 2004

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 9:20 AM
Reply to this message Reply
Here are a few, somewhat unrelated points:

1. I really love the idea mentioned previously of writing Python apps for mobile devices. I think creating a repository of such apps would be a great marketing tool - something like the java applets repositories that surfaced when java started becoming cool.

2. Many people are stuck with an apple to orange comparison.
The comparison with
http://www.python.org/ or
http://beta.python.org/
should be with
http://www.ruby-lang.org/en/
and not
http://www.rubyonrails.org/

3. I don't have a clue as to how this could be done but I would love to see a "sandbox" developed that could be embedded within a browser, enabling someone to write "Python applets" instead of "java applets" or flash programs. This is not the same as using Jython; it is more akin to the concept of vmware's virtual operating systems. If this were to happen, I am convinced that a lot more people would start programming in Python, contributing to its marketing effort in a "viral" way.

4. Perhaps related to 3., having Python usable with the Mozilla Xul framework (this seems to have been in the works forever ...) in a newbie-friendly way. Then create a site with marketing hype promoting using the combination Xul+Python. Such a site could try to compete with http://www.rubyonrails.org/ in terms of "noise" generated. (I know that what I propose is very different from a web framework like RoR.)

5. Start looking again at CP4E - somewhat related to previous posts on the AP lobbying. If 3. above were possible, I think we would see many "cool" applications emerging aimed at various audiences, including children, that could contribute to the promotion of Python.

Chuck Wegrzyn

Posts: 3
Nickname: wegrzyn
Registered: May, 2003

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 10:10 AM
Reply to this message Reply
I have just completed a v1 of a product that I developed for another company. I am now re-writting it in Python; my first attempt. I must say it has lived up to its buzz. My only complaint? I should have done it sooner.

Paul Boddie

Posts: 26
Nickname: pboddie
Registered: Jan, 2006

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 10:56 AM
Reply to this message Reply
A few comments:

This continued obsession with Ruby (on Rails) amongst "major bloggers" is destructive: it "proves" to the casual onlooker that Python isn't a confident choice for certain kinds of applications amongst Python developers themselves. RoR may be good for certain kinds of apps, but the Python community's only flaw here is that it isn't as easy to navigate to a corresponding solution. (I can rant about maintaining some kind of frameworks overview some other time.)

The one true framework is a mirage: for example, I don't want to write my templates in some kind of hybrid Python-plus-templating-language or some non-well-formed deviation from XML, and for many kinds of apps I don't want the object-relational choir singing in my ears. People need a way to identify the best solution for them, not to have to swallow something which is just going to cause problems later on. (If people had wanted to back a horse in some frameworks race, perhaps mod_python, trading on the success of mod_perl, could have been the one to back. The Apache brand name recognition plus hype could have brought people on board, I'm sure.)

The more vocal part of the Python community doesn't appreciate standards: Python would be much less present in the enterprise than it is now if the DB-API hadn't come along. Whilst one true top-to-bottom solution rarely exists in a domain, having some standards does at least give people some flexibility whilst letting them choose.

The point about marketing to developers is quite valid: developers do defy managers and use tools of their own choosing; having some kind of peer acceptance of those tools reduces the resistance that people might have in using Python. If people (even mistakenly) believe that tool X is cool then they will let their peers choose that tool and forgive any subsequent failures, but if they don't have the same belief about Python then the first minor issue that comes up will be used as "proof" that Python is the wrong choice.

As someone said, Python is everywhere. Even if random vocal Python bloggers maintain a tunnel vision fixated on their own specialisation doesn't mean that everyone else is ignoring Python: look at other, broader technology blogs and see that Python is the chosen dynamic language for a lot of stuff. As I've said before, parts of the Python community could do to look around once in a while.

S Deibel

Posts: 9
Nickname: sdeibel
Registered: Apr, 2003

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 12:02 PM
Reply to this message Reply
> I find marketing programming languages to be really
> offensive. If we really want to push Python, we should
> create great things in Python. I'd like to see some books
> like Mark Jason Dominus' 'Higher Order Perl'.


"Marketing" or advocacy can take many forms. The point is more to create great things and then *talk about them* in some forum. The most effective things are how-to style articles, reviews of various choices for a problem (that rate Python highly of course ;-), and screen casts that show something nifty and useful.

We are marketing to people that have a particular problem to solve so we want to provide them with quick easy to use answers to "how do I do X with Python".

We should not (and cannot) try marketing that's more like what was done for Java... that's the stuff that gets offensive because it's not useful to someone trying to solve a problem.

S Deibel

Posts: 9
Nickname: sdeibel
Registered: Apr, 2003

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 12:25 PM
Reply to this message Reply
> > 3) Guido, you must lay the law down and select "the one
> > true way".
>
> Ain't gonna happen. I don't know the application area
> well enough and, frankly, I don't care.
>
> The Python community has to do this without my help.

It would be a real mistake to say there's only one true way. There are simply too many different types of web development to say that one solution will cover them all. The only reason Rails is "the way" in Ruby is that the user base is so much smaller that it only produced one viable solution.

Lone Star

Posts: 7
Nickname: lonestar
Registered: Jan, 2006

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 12:27 PM
Reply to this message Reply
> > 3) Guido, you must lay the law down and select "the one
> > true way".
>
> Ain't gonna happen. I don't know the application area
> well enough and, frankly, I don't care.
Guido, it hasn't happened for years and it won't happen. (Or why should it?) As a BDFL you should be smart enough to take some time (which you now get even paid parttime for) and then make an informed decision. So take a month (or however long it takes) and learn your candidate frameworks (that would be two weeks fulltime) and then decide. Even if it doesn't interest you. Sometimes even a dictator has to do his duty.

Web application development is an crucial catalyst for language adoption and it generates headlines. Otherwise stop whining or posting something like this and watch your troops desert.

Also let me say that I love Python but the web development situation is frustrating to me.

S Deibel

Posts: 9
Nickname: sdeibel
Registered: Apr, 2003

Re: go mobile Posted: Mar 4, 2006 12:36 PM
Reply to this message Reply
> Another area of opportunity that has received a fair
> amount of attention recently is that of IDEs. It has
> become apparent to many that there are no really
> good
IDEs for any of the current crop of dynamically
> typed languages.

This is a good example of how the Python community shoots itself in the foot. Have you actually tried the current crop of IDEs?

http://spyced.blogspot.com/2005/09/review-of-6-python-ides.html

This was also presented at PyCon and was one of the more talked about presentations.

Disclaimer: I work on one of them, but the point here is that this claim is not really true. Sure, some features like refactoring are still missing (well, not integrated yet) but for very many users the current IDEs are quite acceptible.

Andre Roberge

Posts: 6
Nickname: aroberge
Registered: Dec, 2004

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 12:42 PM
Reply to this message Reply
"Daddy, I want to watch Justice League and my sister wants to watch Pokemon. Please come and settle for us which show we should watch, once and for all!"

> > > 3) Guido, you must lay the law down and select "the
> one
> > > true way".
> >
> > Ain't gonna happen. I don't know the application area
> > well enough and, frankly, I don't care.
> Guido, it hasn't happened for years and it won't happen.
> (Or why should it?) As a BDFL you should be smart enough
> to take some time (which you now get even paid parttime
> for) and then make an informed decision. So take a month
> (or however long it takes) and learn your candidate
> frameworks (that would be two weeks fulltime) and then
> decide. Even if it doesn't interest you. Sometimes even a
> dictator has to do his duty.

What if Guido's great instincts as a language designer turned out to be irrelevant/counterproductive when it comes to decide what best features/structure a web framework development should have?...

> Web application development is an crucial catalyst for
> language adoption and it generates headlines. Otherwise
> stop whining or posting something like this and watch your
> troops desert.

I haven't heard Guido whine, nor do I expect "his troops" to desert...

> Also let me say that I love Python but the web development
> situation is frustrating to me.

Thanks for sharing your frustration.

Web development is only a tiny fraction of what Python can be used for. There are plenty of folks out there that use Python as a glue language for numerical work (just to pick an example at random) and who have no use for Web development ability for Python.

Guido should not be expected to make a decision on "what framework is to be used to do X", whatever X might be. In my opinion, you are way out of line both in the tone of your post and your demand.

S Deibel

Posts: 9
Nickname: sdeibel
Registered: Apr, 2003

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 12:47 PM
Reply to this message Reply
> If I were a "marketing guy" (which I admittedly am not),
> I'd define which segments you want to go after, identify
> who the competition is, and build a strategy around each.

That's true if you were a marketing guy working for a large firm with millions to spend. We tried this already on the marketing-python mailing list and it just won't work here. This is a grass-roots marketing effort. Find an area where Python shines and sell that to people by writing about it and showing people how to get started quickly. That's all we need to do!

The difference from the current state of affairs is basically a bit more promotion and a bit less self-deprecation.

Also, sell a solution to particular problems *not* the language itself. People don't switch languages because of the generic virtue of the language, they do it to solve a particular problem they are working on.

S Deibel

Posts: 9
Nickname: sdeibel
Registered: Apr, 2003

Possible focus of energy: beta.python.org Posted: Mar 4, 2006 1:00 PM
Reply to this message Reply
Hi,

I'm working on beta.python.org with a group of about 12 others and we plan to roll out tomorrow. Please look at the site and comment on its quality as an advocacy site (see report website bug link at left). It is not *just* an advocacy site but contains also all the old community, and python-dev content in an integrated design we thought would serve both the existing community and potential new users (we feel two separate sites can be fragmenting and confusing).

Suggestions of specific changes or even content to add are most useful.

After rollout, we'll be looking for content editors -- we would like to assign each part of the site to a team of maintainers that keep it fresh and up to date. If you're interested in marketing/advocacy/promotion please consider helping out with that part of the site (which is currently very small, but could be expanded considerably w/ how-to style introductions). Or participate in maintaining wiki.python.org which has a lot of useful pages for people looking for solutions.

Thanks!

Lone Star

Posts: 7
Nickname: lonestar
Registered: Jan, 2006

Re: Marketing Python - An Idea Whose Time Has Come Posted: Mar 4, 2006 2:05 PM
Reply to this message Reply
> What if Guido's great instincts as a language designer
> turned out to be irrelevant/counterproductive when it
> comes to decide what best features/structure a web
> framework development should have?...
It is not about choosing the technically best. It about choosing something that is ok (not the worst) and allowing one solution to gain enough momentum and maturity. And this is something that hasn't happened in the past.
>
> I haven't heard Guido whine, nor do I expect "his troops"
> to desert...
It wasn't Guido but the original author. Can we assume that somehow Guido's opinion is the same when he posts content from other authors? The original author complained about the hype around ruby and rails and python not getting the same hype. I personally know three people coming from python trying rails (because of the lack of a python alternative), liking rails, beginning to like ruby, using ruby more and more, using python less and less... That is how you start losing people and it is something happening in reality.
>
> There are plenty of folks out there that use
> Python as a glue language for numerical work (just to pick
> an example at random) and who have no use for Web
> development ability for Python.
Well, I would say there are more people that would like to use python for webapps than people using it for numerics (but that might be subjective and wrong).
> In
> my opinion, you are way out of line both in the tone of
> your post and your demand.
You shouldn't be too picky about my tone since I am a non-native english speaker. But Guido's "frankly I don't care" doesn't do anything to better the situation... Au contraire, it plays in the hands of the people trying to spread FUD against python: "If Guido, the creator of Python, has no interest to use python for webapplications like he stated several times, why should you beloved customer invest your money in a python based webapplication. Try Ruby on Rails, everyone likes it, even Matz the creator of ruby". Customer lost!

Joao Pedrosa

Posts: 114
Nickname: dewd
Registered: Dec, 2005

Re: Possible focus of energy: beta.python.org Posted: Mar 4, 2006 2:12 PM
Reply to this message Reply
Another new Web Framework in Ruby:
http://pandora.rubyveil.com/pandora/Pandora/Introduction/Pandora

Look at the web site, and how pretty it is. The Framework sucks in the way that it is double licensed (GPL/Commercial), though the prettyness of everything may be worth it. :-)

I mean. Let's go back to culture. Ruby has had great influence from the Japanese culture, because most of the core development of Ruby has been done by Japanese. Japanese seem to be very good at C and such low level stuff. Matz, on the other hand, gave this low level stuff a touch of high level constructors, mainly by looking at what other languages were doing. Once more some may say that the Japanese culture may have influenced this "copying" of ideas, but all people try to copy, ones more than others, ones better than others.

Ok, now let's get Rails in the picture. Rails was created by someone who was about to get degrees in Business Administration and Computer Science (I think, forgive me if the degrees aren't correct). On top of that, this someone was using his Mac to develop Rails -- talk about bridging worlds. And Mac people are fanatics and all that, right? They like their stuff. By looking at web sites that Mac people create, you may notice that those web sites tend to be different. They tend to be less cluttered than web sites created by people on Windows. I don't know why is that. Maybe people try to copy stuff they find in their environment, and that means that Windows people will copy what other Windows people do, and Mac people will copy what other Mac people do. So, take the Apple.com web site for example. Does it influence how Mac web designers create their own web sites? For sure. Does the MacOSX desktop influence how Mac GUI designers create their GUIs? For sure.

Ok, so, Rails brought the great web designing that happens in the Mac land, the great dynamic programming that's possible with Ruby, and other stuff, together. It's a sui generis mix. There's no way the Python land can copy that. Python is traditional on Windows. Python is traditional on Linux. That usually means hard core programming, that is, the kind of programming that I like. But that's not what most people are after. People are after easier web development. Not only that, nowadays people want the "easiest" web programming that they can get, and that means that it needs to be popular and self-sustainable whenever possible. It means that dual licensing sucks. it means that GPL sucks. It means arbitrary constraints suck.

Don't try to copy Rails. Try to beat them in their own game, and that means being the best in web programming.

Flat View: This topic has 117 replies on 8 pages [ « | 1  2  3  4  5  6  7  8 | » ]
Topic: Marketing Python - An Idea Whose Time Has Come Previous Topic   Next Topic Topic: Reference Architecture


Sponsored Links



Google
  Web Artima.com   

Copyright © 1996-2014 Artima, Inc. All Rights Reserved. - Privacy Policy - Terms of Use - Advertise with Us