[Typo3-dev] Re: The future of typo3

[Michael Zedeler]

Hi Kasper				

> > I
> >strongly believe that the content-publishing-paradigm is dead
>
> I strongly believe that user interaction is dead. I browse websites, thats all that makes sense most of the time. Thats what big corporations seems to think as well. I never use community sites like those based on PHPnuke anyways.

This part of the discussion cannot be finished by realizing that one of us is right. I think that the answer is that there are people with different needs. I take that you want typo3 to be in line with the "we publish - they view" paradigm. Fair enough. Of course - somebody out there will be using it.

> >Please enlighten me -why is it such a good thing to separate the backend
> >and the frontend?
>
> TYPO3 has won recognition as a web-publishing system. Having a separate backend allows it to do other things - eg. work as a totally backend-only system. In fact the whole frontend page is encapsulated in the "cms" extension and therefore is strictly not a part of the core.
> I wouldn't like having 50000 users in my user table and only three of them being backend users anyways.
> The backend is a convenient "system" for all kinds of administrative purposes - the frontend is a website, totally different. Well, I'm not going to argue this - to me it is just obvious that each part has its right place. I do agree that some sharing of code would be nice for some parts - like editing for frontend users.

Fair enough - if the descision is to stick with the paradigm above. I won't work with a system like this because that the websites that I develop require more flexibility and less strict separation of user roles.

> >More extensions aggravates the problems that I am addressing. All the
> >code needed for front end editing is already there. It just needs to be
> >*enabled* for front end editing.
>
> Its easier and safer to just re-write a frontend version. Two versions with separate access schemes is not too much redundancy.

So merge the access schemes!

> All the "bridges" between front end and
> >back end users (front end user - to back end user mapping, backend login
> >in front end and so on) are symptoms of poor design.
>
> Right.
>
> Mixing frontend and backend users are symptoms of a poor design as well, a design I wouldn't use if I had a very important website to run.
> I once studied electronics. There you have various devices to transmit signals with "hermitical separation". Why? To make sure that the risk of error is very very low under extreme unforeseem conditions.
> The same here: Having two user tables in design provides a safer system.

I don't take that argument. If the system had some sort of basic, simple and well tested security module that handled the databse communication, it would not be an issue. I think that the situation now is far more dangerous because some of the publicly available extensions may be open to SQL-injection attacks due to the fact that there is no database abstraction layer. Such a security hole could expose the entire database.

> >Se my comments above regarding symptoms of poor design.
>
> Michael, why not present us your perfect design? And then, lets see if it stands for scrutiny better than TYPO3 or if it only expresses what you happen to think it best?

I have already suggested merging the frontend and backend user and group tables. Those are the only suggestions that I have for now.

> >good, a mere reconfiguration should make it possible to emulate a wiki.
>
> Is it OK to make software which cannot emulate a wiki?
> Strange idea.

This was taken out of context. I had three other examples: intranets, extranets and e-commerce.

> >In my opinion: that is because that you have worked out a way of using
> >typo3 already, and youøre not able to see how inconsistent the parts
> >that you have gotten accustomed to avoid really are.
>
> Based on my general impression of you from earlier mails (which could be badly mistaken of course) you are also filled up with preconceived ideas from other technologies which makes your opinions just as subjective as ours. So why are you more right?

I am not nessecarily more right. I am trying to convince you that I am right. Not by writing "I am right and you are wrong", but by using arguments. I believe that is a fair way to discuss and test existing and new ideas. I am open to realizing that I have been wrong in my assumptions. The fact that you want to stick to the "publish - view"-paradigm surprises me, but I respect it and it settles many of the questions I have.

> I see you as a guy who will have a hard time with TYPO3 because you have to fight these preconceived ideas for a time and you will hate TYPO3 for not fitting the way you think. But there are others who get it more quickly, probably because they think the same way or otherwise already work in the paradigms TYPO3 does.

I am trying to understand what the mission of typo3 is, and I have already learned a lot. From what it looks like for me, typo3 is going in a direction that I wouldn't take it, but that is fair enough. I might use it in the future - and after talking to you guys I know what tasks it is intended to support and - especially - what tasks it is not intended to support. That will make me more confident if I ever was to recommend using typo3.

> We have had people like you around before. Thats just fine with us. But we also recognize that TYPO3 doesn't fit everyone (and we don't intend it to do) and if so, please don't waste your time here. It doesn't help you, it doesn't help us.

If this is what you want, I'll stop posting after this mail. ...and you can go about your business with typo3 as usual.

> >And away goes the famous separation of code and layout.
>
> Hardcode HTML is beautiful. I have just implemented a big project in france using this method and we have been very lucky doing so in order to get the job done in time and in the best way with best performance.
>
> Could it be that you have been accustomed to this paradigm of separation, never questioning it?

Doesn't the cache leverage the performance loss induced by using templates?
I'd be happy to discuss pros and cons of code/layout separation.

> >I think it is more appropiate to give such a document as "Getting
> >started with typo3" an overhaul.
>
> Strange comment: That is the MOST RECENT document and it is all the other documentation that is REALLY POOR. If "Getting Started" is really that bad, there is no hope for the rest.

Please take a look at my previous suggestion regarding documentation. You should do something actively to support people in making money on documentation. Sorry  I was wrong - the document that I was writing about was "Creating a basic extension". It lacks information about how to let the front end users edit data in the system using the plugin created.

...but the again: if you don't really believe in this sort of interaction, maybe it is okay not to write anything about it.

> >The principles behind the extensions are really good.
>
> At last, something good...

Yes. The extensions are nice. This is one of the things that I think makes a great difference between typo3 and its competitors.

> >What I would like
> >to see is transferring more functionality into extensions and cutting
> >down on the core functionality.
>
> You should have seen TYPO3 v. 3, then you would appreciate what we have got not.
> But have you actually REALLY got into what is core and what is not?

Probably not. The source code is overwhelming.

> >Fair enough. But the back end should be an application written on
> >exactly the same foundation and with the the same tools as everything in
> >the front end. My point is not that I don't like the backend. What I
> >don't like is that it is being treated as a completely separate
> >application in every respect all the way down to the very core of typo3.
> >Thats not nice.
>
> Oh, thats very nice. I just LOVE it!

Very funny :) And a nice way to avoid responding to the suggestion above.

> >Please don't be offended. I am writing this to help. It may sound harsh,
> >but I _will_ respect arguments against the standpoints that I have
> >presented. It is clear that typo3 is your darling (a darling that need
> >not be killed!). But I think that too much "I know whats best because I
> >just know" causes degeneration of the system. Its a good idea to start
> >discussing why it is so great and what can be done (even) better.
>
> I am to some degree offended. I admit that. I realized after the other mail that it was primarily because I agreed with you on many points but had just gone through a process with myself where I had to accept that the best thing would be to focus on what IS DONE instead of wishing to trash everything and do it all over - which is something living only in our imagination still.

Trashing the system is extreme. I suggest some sort of refactoring that will let developers reuse much of the code that has already been developed. It is hard for me to imagine that you think a lot about all the stuff that has already been done - after all - you are the head of typo3 development. I believe that many of the suggestions that we have discussed will make coding easier for everybody - including the cores developers.

> I chose what is most wise I believe and what carries most responsibility with it for the thousands of users of TYPO3 at the moment: To focus on what we DO HAVE. But all your suggestions just reminds me of my lack of time and of the dream which had to be put aside for the sake of reality.

I know that game all too well. I have no part in the responsibility that you mention, so I am just ranting on about how nice it could be if we had 90348290843 years to improve the system :-)

> That is also why I have kindly directed you to the team leaders of various project with your suggestions and also implying that real help only comes with a donation of time/initiative or money. Its easy to just mention what should be done differently.

I know. But this discussion is important to me, because I get an insight into what I would be investing in, if I did. An insight that I am very happy to have now.

> Anyways, I actually do think that "I know whats best because I just know" - but I also listen to arguments. Some of your rather detailed suggestions to me has not been bad - I just knew this was not going to be (with me at least) and so I didn't spend too much time on a detailed conversion on them. But I did listen.

Great.

> >I have been busy making money and running companies.
>
> I had that option as well back in time. But decided to do TYPO3 instead.

I regret that I didn't do the same. But hopefully I have many years left on this planet, so that might change.

> >> So why bother giving us advice?
> >>
> >Because I think that it was the best thing that I *could* do for the
> >project.
>
> ok, but still, why bother? If all the principles are wrong? Principles are the core of systems, most of them not to change.

Because not until now have I understood that what looks like random design descisions acutally forms a pattern that is in line with something that I disagree with. And you have confirmed it :)

Regards,

- Michael Zedeler (mzedeler)

-----------------------
This thread is located in the archive at this URL:
http://typo3.org/1427+M5edb97ccbde.0.html