[Grok-dev] Re: maintaining the Grok website

Sebastian Ware sebastian at urbantalk.se
Sat Sep 15 19:35:55 EDT 2007

15 sep 2007 kl. 22.39 skrev Martin Aspeli:

> Hi Sebastian,
> First off - sorry if I came off a little strong. I certainly don't  
> wish (or have any right to) criticise what you're doing. I do hope  
> that the Grok community is willing to choose a system that is  
> *appropriate* for their needs, rather than insisting on creating  
> something from scratch for the sake of having it in Grok. If there  
> is an appropriate Grok-based CMS, that does of course sound  
> attractive. But if getting there will take a lot of time and work  
> and introduce a lot of uncertainty, it seems a bad choice to me.

I am a pioneer by nature. I see that more as an opportunity :)

>>> Sebastian Ware wrote:
>>>> I think it would be very valuable to have the website run on a   
>>>> Grok  based CMS. We desperately need show case applications. On   
>>>> the other  hand I can see the benefit of a mature CMS such as  
>>>> Plone.
>>>> However, being able to boast that we are so confident in our  
>>>> own   technology that we even run the Grok website on a Grok  
>>>> based   application is worth a lot to those evaluating it.
>>> *shrug* - I think you're overplaying this card. It may be a neat   
>>> thing to be able to say, but it's not going to be the clincher.  
>>> All  kinds of lame PHP projects run their websites on PHP, that's  
>>> not a  good reason to chose them. :)
>> Shrug or no shrug. I humbly suggest that you are a bit off the  
>> mark  by comparing Grok to "any lame PHP project".
> :-)
> Of course, I love Grok, but that's besides the point.


>> Rather, consider if Zend- corp ran their site on .NET.  I think  
>> that would really make some
>> difference. Even if they ran it on PHP4 it would signal that they   
>> weren't 100% commited to the latest versions of their framework  
>> and  engine.
> I don't think it's a directly comparable situation. Zend may be  
> vested in PHP - that's the language. Grok is higher up in the  
> stack. Plone may be vested in the Plone CMS. Grok is not a CMS,  
> though. And Plone runs on Zope, which shares a lot of  
> infrastructure with Zope 3.
> The pragmatic question here should be, what system gives us the  
> best tools for meeting our web-site building needs. And "best" here  
> must include some notion of support, maintainability, and proven  
> stability.

I agree in most. But then again, proven stability is kind of saying  
that anything that hasn't been around for a while is out of the  
question. So anything based on Grok would be disqualified out the box.

I have personally designed and been involved in one of the most  
extensive scalability test performed on the .NET platform. The  
purpose was to evaluate how well object-relational mapping scaled. I  
agree with your concern, but the scalability and stability issue is  
on the table for every Grok app out there. If we have to look into  
this, there is an opportunity to learn more before the adopters are  
stuck with the problems. This is both good and bad.

>> Change history is good, but I don't see versioning or staging as  
>> an  important feature for the Grok-website. What Grok really needs  
>> is a  really lightweight solution that has a smart knowledge base/ 
>> commented  API kind of thing. So I am guessing that a lot of the  
>> benefits of  Plone are overkill at this point.
>> Lets keep it simple :)
> It's not simple to maintain something yourself. I guarantee you  
> that you're going to need or want more features that you initially  
> foresee. You're going to end up re-inventing things that are  
> already in Plone. I don't buy that it's "overkill". That word has  
> lots of negative connotations. It's true that you may not need  
> every feature of Plone, but that doesn't mean it's not well suited.

I agree. But I am bit more into the agile idea of "lets cross that  
bridge when we reach it".

> Also, Plone has spent a lot (!) of time on usability. Even if most  
> people here could put up with something that wasn't very end-user  
> focused, they're still going to be more productive if they have  
> tools that are easy to use. Getting solid, good usability takes  
> time. I haven't seen your CMS, so I can't comment on it, but I know  
> that I've never seen a brand new system, written by a single  
> developer, that got the usability right on the first attempt.

I know what you mean. Seeing is believing. I agree with that 100%.  
I'll rest my case when Luciano says what I sent him sucks... :)

>>> If you have to maintain your own CMS "for the sake of it",  
>>> you'll  take away time that you could've spent improving Grok or  
>>> making  other, more impressive demos. I think there are lots of  
>>> other types  of applications where Grok really can help you build  
>>> best-of-breed  solutions - why compete with Plone head-on?
>> I doubt this would take away time from making other more  
>> impressive  demos ;) I also don't think you should feel worried  
>> about a  lightweight CMS based on Grok competing with Plone.
> I don't.
>> I would rather say that this could offer adopters of Grok a   
>> lightweight and easily customisable CMS with a fully functional  
>> back- end that could quickly get them up to speed with a certain  
>> type of  project. This would lower the barrier of entry by an  
>> order of magnitude.
> Possibly.
> Are you also promising to maintain this CMS, fix bugs quickly when  
> they're discovered, make sure that it scales, and keep it up to  
> date with the evolution of Grok?

A small code base is obviously a lot easier to maintain than a large  
one, and if one doesn't have to worry about side effects sometimes it  
is a lot easier to roll your own, but this is definately a valid  
concern. Whatever it is, it certainly won't be mature in that sense.  
But it might nevertheless do the jobb good enough. :)

I think you should keep thinking about the Plone solution. If my (or  
another lightweight Grok based CMS) turns out to be up to scratch  
within the next week or two I think it should be given a fair shot,  
but otherwise I don't have any personal objection to idea of using  
Plone 3.


>> I don't know the status of the other CMS' based on Grok. I have  
>> sent  my code to Luciano because I am confident that he will make  
>> a sane  evaluation of what use it might have. :)
> Martin
> -- 
> Acquisition is a jealous mistress
> _______________________________________________
> Grok-dev mailing list
> Grok-dev at zope.org
> http://mail.zope.org/mailman/listinfo/grok-dev

More information about the Grok-dev mailing list