Game over. We lost. Nothing to see here, move along.

Aaron Trevena aaron.trevena at
Fri Nov 18 15:11:04 GMT 2005

On 11/18/05, Ben Evans <ben at> wrote:
> On Fri, Nov 18, 2005 at 01:54:35PM +0000, Aaron Trevena wrote:
> > On 11/18/05, Ben Evans <ben at> wrote:
> > > On Fri, Nov 18, 2005 at 11:31:04AM +0000, Dave Cross wrote:
> > > I've been meaning to write up my concerns about this topic for a while
> > > now, but have been sitting on my hands and waiting for my initial
> > > irritation to subside.
> > >
> > > I've now tried three times to get a Maypole app up and running and have failed
> > > each time, taking 3-6 days of wasted effort each time.
> >
> > You never thought to email the list?
> What list? I looked for a suitable one and couldn't find anything active. and

Both linked from the wiki, both with 60 - 100 mails a month .

> > Or add anything to the wiki about your problems?
> ? Persistently down, unusably slow when up and full of wikispam
> and broken links.

I've noticed it down very occasionally, not slow and the wikispam has
been cleaned out regularly (we plan to move to maypole based mitiki in
the next couple of weeks which will resolve this).

> > Or did you report any bugs to
> How does one report a bug when the test application for something like a
> framework won't even come up? That's a deep integration issue, not something
> one can fit into RT. And what sort of resolution is it going to get, other
> than WFM ?

Looking at the help provided on both cdbi and maypole lists, I'd
expect something pretty helpful pretty quickly.

But as Maypole uses CDBI out of the box and it's support for sybase
isn't something I'm too hot on that would be the obvious place to

> If you can find me some examples running off a Sybase DB, I'd be very
> interested in seeing the source.

Nope, db connectivity is closely linked to the ORM you use, so most
support is on the relevent list. But you'd still get people happy to
help where possible on the list if you'd bothered asking.

> > > [ several days of banging head against wall snipped ]
> >
> > Sorry, but I'm by no means a great hacker, but wtf? I never had these
> > problems, even when I was stuck on a train with no internet
> > connection, and no reception on my mobile to phone a friends I managed
> > better than that.
> That's precisely the attitude that made me not want to report problems
> on the previous two attempts.

Wuh? What kind of response do you expect after ranting about your
experience on a desert island where you couldn't possibly talk to
anybody else about maypole?

When I have genuinely been in a situation where I couldn't talk to
anybody I was able to understand what was going on. But then I know
the underlying technology and was using stuff known to work with it.

> Maybe if you tried to be a little more conciliatory to people that haven't
> drunk the Kool-Aid, you might get a few more users.

There is no kool-aid, there are just people who have bothered to talk
to each other about stuff, instead of running off to an unrelated
mailing list or blog to bitch about it.

> Maypole 2.10 - about 3-4 weeks ago. Latest versions of absolutely everything
> available (and all previous versions, so I can use earlier versions if required).

Right - so subscribe to the list or at least look at the archives -
also check your basic application works with plain CDBI first. You did
try doing some simple test code using the underlying ORM with the
schema when you encountered some problems right?

> > Or you could, and obviously you would rather avoid it -- speak to
> > people about this stuff rather than bottle it up for a year and vent
> > long after anybody can help, and without any useful details of
> > specific problems.
> Hardly. I had a deadline to prototype an application and make a technology
> decision. I'd deferred it because I had heard great things of Maypole and
> estimated I could get a workable POC in a few days effort, and because
> I knew that I had a fallback (albeit at the cost of some supportability
> and maintainability) of using my homegrown framework.
> I found Maypole to be unusable and bordering on abandonware from all the active
> participation and help that seemed to be available.

You mean, the entries in rt that are shown to be fixed in SVN, or with
workarounds or active discussions of the problem, or the irc channel
that has been active for a month or two, or the mailing list with
public searchable archives that is easily found in several places on
the first page of google results for 'maypole mailing list'?

Really -- there is a shedload of support if you bother to look. I
don't mind or care if people say something is broken, or doesn't work.

I do get annoyed when people spend their time providing support on
mailing lists, updating the wiki, working on new documentation and
being on irc to help newbies and I get this bullshit that none of it

> I'd be prepared to give it another go, but it will have to wait until I
> have a project which can support another week's worth of potentially
> wasted prototyping time.
> FWIW, I suspect that the largest irritations I suffered were more to do with
> CDBI not supporting Sybase properly and that Maypole was simply not being
> smart about showing me where the errors were really coming from.

Cool. You know that would be a really handy bug report.. if only there
was some way you could report that didn't involve the impossible and
clearly out of use mailing lists, request tracker, or irc channel?

> The Maypole documentation still sucks, though, and really lacks a decent case
> study for "we have a database, now lets code up some maintainance screens
> in no time flat", which seems to me to be an excellent primary use case
> for the whole web frameworks concept.

Like the BeerDB or BookDB - to be honest, we could do with working on
both of these things, but there is a shedload of new documentation in
SVN and that is one of the reasons for the delay in releasing 2.11.

Also we're working on some more Maypole applications to try out and
look at how they work. Dave Baird is doing another example database to
web application containing design patterns, that should be helpful.



More information about the mailing list