Anyway -> Performance is a key argument for sure! A possible performance leak of cake is that they use much more functions by default. Django let the programmers do more work on there own. Whats clear -> a function call does take more time than just spid out a var. i think a proper test without usage of functionfields and scaffoldering would make this difference much smaller.
Okay, functions cost performance, BUT they do a lot for you and your fast development. If you want to have insane performance you leave out any framework (except a SOA or cacheing framework). We get now in a fine granulated state -> what do we want to have from a framework. performance over lazyness? how heavy is this impact to the project. should we comepile the project as binary? which ORM methode works best and most effectiv.
I think that performance strongly depends on lazyness for the programmer. You can tweak everthing away if you want to but than you loose again the maintainability. More abstract less performance. Iam unsure if cake make use of the new namespace function from php. I dont think so, but thats another point which defently makes PHP a not so good choice. i seriously love python, but in my oppinion abstraction over performance (and in the normal world -> python is much better for abstract structures than php).

Its a bit like the Assembler discussion. People who did a lot with it, would only move slowly to a more abstract language, cause they miss performance. This argument is always valid, but never without loosing maintainability.
And than we got the „how low do i have to go“ question. Do i need to programm the HTML response by myself to get rid of the Apache? The answer is NO! Performance is important but not with the risk to loose security / reduce stability.

About the Convention over Configuration ->
Thats the best if you need to work in a team. Without a question its harder to understand in detail, but the reusability is awesome (even if the creator is not that good). If the „eat your own dogfood“ (use what you have written once, even if you dont like it anymore. And if you dont like it anymore -> redo this part until its okay again) rule is done refactoring will be the next step until the component reach a level which is good enough to be used in all projects.a

http://www.google.com/trends?q=django+python%2C+cakephp+php&ctab=0&geo=all&date=all&sort=0

Simple search not for django python vs cakephp, but against cakephp php, so you have the language embedded to both searches.

With that, you see a much more realistic result, since both can win/loose by there language.

As you can see, cakephp looses a lot by adding php, so we can guess that this is the same for django with python.

About both frameworks:
I tested alot with cakephp, and worked (after that) with django. My result: CakePHP seems to be easy, but has a lot to much Convention over Configuration, somethimes I couldn’t understand why cakephp does something. I need to remember to much logic.
Django on the other hand, while still using a lot of magic, forced me to do some important things in configs and definition which really helps me to unterstand what django does. I found it much more easy to understand complex examples in django then in cakephp.

Also, django seems to be MUCH faster then cakephp, in my tests up to 20x faster. Other benchmarks I found in the net says the same (but allways with the info: possible that there is room for a lot of optimizations in both frameworks)

Don’t focus yourself to much on the lang if you have a bigger project. Focus on the base of functions you want to have. Predefine your needs befor you choose the framework. If both framework fulfill your needs -> choose your prefered programming language (if performace doesn’t matter).

This link shows a different story

Wenn die Seite überarbeitet wird und mir andere User über diese änderungen berichten, werde ich sicherlich dem ganzen nochmal einen Blick gönnen. Aber momentan bin ich 2 mal auf der Seite gewesen (Registierung, Wochenlange Pause, Anmeldung) um frustiert von der Seite zu gehen (jeweils pro besuch ). Ich lasste es auch keinem Developer an, dass die Seite so ist, wie sie ist. Diese entscheidungen sind wahrscheinlich an einer anderen Stelle getroffen worden (hoffe ich jedenfalls, weil wenn ein Pageentwickler wirklich nen login „manuell“ erstellt muss, wäre das schon etwas was man als „basics“ bezeichnen würde).

Ich will auch jetzt nicht noch die Fähigkeiten der entwickler angreifen, weil es wahrscheinlich zu unrecht wäre. Die planung der Marketingmaßnahme war jedoch für die Tonne und als gefrusteter User muss man auch dieses mal klar und deutlich an die Macher kommunizieren. Ansonsten können sie nur ihre Statistiken anschauen und sich wundern, warum der Erfolg ausbleibt.

da waren auf jeden fall experten am werk

vielleicht wirds ja noch

bevor man ABTAUCHT mit solchen kommentaren sollte man noch eine chance geben

for sure my view is influenced by my personal likes. I like to have describe all data for the query in on array instead of multiple functions. Thats against the Object Orientated Design Pattern, and i know that, but i still like array way
Its just faster to see, why the Setting XYZ is in that query.

Django has got some massive features which i rly like (SQL Alchemy) and Python is just a awesome Language (especially cause its strongly connected to the OS. So more specific tasks like ‘grab a thumbnail from a video’ doesn’t need to „exec“ a program to do so) and thats the big plus of Django.
But as long as we don’t care that much about the language (depends on the needs of the site), its just the questions -> what does the job faster / more intuitive for yourself.
In my oppinion django is damn fast if you have a „typical problem“ (like user administration, or the comments on blog entries) but if you leave that are it feels like -> django was build to do that kind of tasks and than you need to do everthing else on your own.
Whenever i use Django, i feel forced to typical problems. If you need something changed you need to override everthing until you have rewritten the complete part which you wanted to use. Dont get me wrong, i see the effect of djangos speeding up the development, but its seems to „focused“ on this basic tasks.

And if we talk about the Admin section -> Yeh, its awesome work. It looks good and feels great to use! But if we look at CakePHP -> Doesnt Scoffoldering doesnt the same. Maybe it doesn’t look that good, but its does the same things (CRUD). The routing of the URL’s are done by just delete on char in the Route.php and with the bake command you get all files written to the file-system if you want to alter them.
Don’t get me wrong again -> the Django Admin Panel is really nice, but „is it really that awesome“, that you leave other features behind?

Without a question it’s a closeup fight between them, and so I’m really thankful for every comment with another individual opinion. (hope to get some PRO Cake voices too, but maybe Cake users don’t care enough about other solutions, what would be sad. There is no „100%“ correct answer, its all a question of likes and dislikes and topic specific).

—-zip — and just because its on my mind atm -> one big pro for django / python is mapnik (python based mapping solution). its damn good, and there is no other free solution which is that advanced (big G. solutions doesn’t count -> its not free, so its illegal ).

[edit]
and greets to Australia! Have been there 2 years ago, where if found some Django loving friends too (they got django as a topic in the university, which is definitely a good step -> learning the basics of a language is good, but learn to use the basics productive is even better)
My youger sister does „work and travel“ in OZ atm too.

thanks for this great comment !

in front : your right. i need to rework this article. mainly its not well formed and spelling is bad, too. i will try to get it more visual (its just so hard to do with „A is better than B because of XZY“).

Anyway,

about the stuff you pointed:
for sure django has most of the possibilities of cake (and in some parts a lot more, like mpeg decoding and all work which is related to direct system access). The opportunities which python has against php are more or less clear and I’m trying to see the languages on a equal level (what is maybe impossible python wins this one without a question)
Lines of code are for sure a good indicator for measuring a framework, but it always depends on the knowledge of the person who coded with the framework. I don’t think thats its hard to mess up a page with both frameworks. Just ignore the function which are already there and your code will double. In my opinion you have to calculate in that you rewrite the pagelayout and ever box which is displayed (cause otherwise all pages of a framework would look pretty similar). So i dont care about scaffoldering and generic views that much. They are for sure a good base but in the end you need to rewrite most of the stuff to fit your needs.

About the google code pages: in fact there is a huge amount of projects which are based on / extend django. One possible reason could be that cake has got its own „cakeforge“. Ive tried the search to, and your right. With „cake“ as search term you get some pages but non of them are really related to CakePHP. I’ve tried CakePHP as search term and it was a bit bitter but still lousy.
The other possibility -> most CakeUsers rly use the cakephp.org page and nothing else (maybe because they don’t know the pages or they just don’t need them).
What i wanted to say in the article -> bring all related stuff to one page is good and i like it. Django just don’t care that much about the community pages and let them work that out alone (which is totally okay, cause most of the projects does it that way), but corporate identity and the same look and feel just let you get easier into a new page. But to say something bad about the Cake Page: The constantly Donate requests are a bit to much (okay, i’am drifting away ).

About the Tickets-> nope, i’ve never fixed a ticket for CakePHP or for Django. This view is totally from the outside.