[ZWeb] "Token 'ATOM' required, 'EOF' found" error in zopeorg

[robert]

Jens Vagelpohl wrote:

> On Jan 21, 2005, at 15:47, Andrew Sawyers wrote:
>
>> Chris Withers wrote:
>>
>>> Andrew Sawyers wrote:
>>>
>>>> In the cases where I've had to catch the query prior to passing it 
>>>> to the catalog, one has the opportunity to inform the user.
>>>
>>> Well, if we could get zope.org onto 2.7.4 or some such, you can just 
>>> catch the requisite exceptions in the python script that calls 
>>> ZCatalog.
>>
>> Well, I was talking to Jens day before yesterday, I'm sure there's 
>> several capable people who'd be more interested in helping with 
>> zope.org if it was just a simple CMF site with skin to make it 
>> pretty.  Maybe it's time for us to reexamine things?  None of us have 
>> the desire or time to sludge through what's there now.  At one point, 
>> Jens and I had a completely migrated zope.org (the old one) onto a 
>> straight CMF based site.  Now who knows what's there; it's not even 
>> on a CMS anyone at ZC uses on a daily basis (i.e. it's based on Plone 
>> which adds several layers of unecessary indirection) and we don't 
>> have people who are familiar with it chomping at the bit to help us 
>> out.  This is a huge roadblock IMNSHO.
>
>
> Amen to that.
>
> It's not just the fact that there is other software that adds 
> unnecessary layers and that no one at ZC uses, unfortunately those 
> layers are also badly done and badly broken in way too many places.
>
> The current maintenance strategy is based on the hope that a lot of 
> fixing will some day make the software do better, but IMNSHO I cannot 
> see that succeeding. There are too many problems and too many 
> fundamental flaws in the design and implementation.
>
> While speaking with Andrew I have volunteered to help create a saner 
> basis for zope.org. It is unlikely that ZC is willing to spearhead 
> such an effort (no time, no willingness to throw good money after bad, 
> etc), which I understand. The biggest problem that creates is a 
> leadership vacuum and any efforts would soon peter out in a flood of 
> discussions from a whole lot of more or less well-meaning people. It 
> cannot work without a tightly organized *small* group of developers 
> that can work without interference from the community at large. That's 
> how things get done.
>
> jens
>
> _______________________________________________
> Zope-web maillist  -  Zope-web at zope.org
> http://mail.zope.org/mailman/listinfo/zope-web
>
If anybody does take the lead, we at redCOR (and not only me) will be 
happy to help on implementig.
Robert