[thelist] content management systems?

martin.p.burns at uk.pwcglobal.com martin.p.burns at uk.pwcglobal.com
Thu Feb 21 10:59:01 CST 2002


Memo from Martin P Burns of PricewaterhouseCoopers

-------------------- Start of message text --------------------



To:   thelist at lists.evolt.org
Subject:  Re: [thelist] content management systems?


>While brining a highend solutions like Vignette into the picture, you're
>immediately bringing the cost of the project to $100,000 US to $500,000
>before integration or training even starts.

True, but
a) the original question was about that end of the market (Interwoven is
   about the same kind of price), which is why I tuned what I said to that
   level. If you're working at that end, and suggest a home rolled solution
   then you just wouldn't get taken seriously.

b) I quite carefully talked about packaged solutions in general, then gave
   some examples at the high end of the market. My points would equally
   apply to Zope which is free (as in beer and speech), or to Adrian's
   QuantumCMS which is much cheaper than the sums you mentioned.
   Even Allaire were offering a complete system, including boxes and
CFEnterprise
   licenses for around 80kUKP IIRC, and with good purchasing people,
   you'd get away with not paying licenses for DR, staging or development
   environments.

>Yes, we're talking about a corporate solution, but almost 2/3's of the
>companies that I've seen convinced into purchasing a license for Vignette
>StoryServer, Blue Martini, Interwoven, or any of the others were also
>convinced they needed a brand new server to run it, licenses for Oracle
8i+,
>and more... when clearly this was an overkill and cost a cool half a mil.

If you're in that end of the market, you're not talking about a personal
home
page, you're talking a robust system. And your description is what you'd
be looking at for a corporate robust system. If you're genuinely in that
market,
then that's cheap-ass - you'd spend that on a potentially throwaway pilot
because the learning you'd get out of it is worth more.

If you're really in that market, a rollout is going to cost a million a
month,
maybe, over a year (and that's pounds, not dollars) plus if it's an
external
facing system, maybe ten times that on marketing and business support.

Now if you're being sold a corporate robust system and you're not that
kind of client, then that's a failure in capturing the business
requirements -
which will be the case whether it costs 500k or not. The system is
overkill,
not necessarily the money (they're separate issues).

>Solutions like these do make sense in some situations, especially when
>connecting to legacy systems or for massive publishing needs (I worked
with
>FoxNews to integrate Vignette).  But for the majority of instances out
>there, I've found that "home rolled" solutions on a VA Linux server using
>PHP and MySQL sufficed.

Homerolled will nearly always end up costing more for the same performance.
Really, how often do you need to code a security system for paying clients?
If you're doing a bespoke system each time, then you really are wasting
their
time, their money and storing up substantial problems in the future when
your
handrolled solution just doesn't work any more and you *then* need to buy
a real system.

It comes down to the same question I askeds in my earlier post:
>> Other than the personal vanity, why would you want to do that for
>> yourself? And why would a paying client want you to develop it from
>> scratch?

Cheers
Martin


----- Original Message -----
From: <martin.p.burns at uk.pwcglobal.com>
To: <thelist at lists.evolt.org>
Sent: Thursday, February 21, 2002 10:47 AM
Subject: Re: [thelist] content management systems?
> >so it's better to
> >roll you own tool using something like zope which lets you "construct"
> >an app. using differents building blocks.
>  Nearly all grown-up CMSs do this - they're toolkits which you need to
> layer stuff on
> top of which relate to your specific needs. And if they're half-way
> sensible, then you
> can define your own building blocks either from primitives (right the way
> down to
> Java/Python/Perl/VB/whatever code) or from their provided composites.

> Most packaged CMSs and linked packaged products (eg Vignette, together
with
> Verity, as a front end to an ERP like SAP to pick a high-end product set,
> and
> similar bundles could be put together at a lower end) now can do*enough*
of
> the things you'll need for nearly any purpose (and almost *all* of the
> supporting
> functionality like searching, user/security management, workflow etc)
that
> unless
> you want to compete as a vendor, you just wouldn't ever write one from
> scratch
> for a client.
> Other than
> the personal sense of achievement, why would you want to do that for
> yourself? And why would a paying client want you to develop it from
> scratch?


--------------------- End of message text --------------------

This e-mail is sent by the above named in their
individual, non-business capacity and is not on
behalf of PricewaterhouseCoopers.

PricewaterhouseCoopers may monitor outgoing and incoming
e-mails and other telecommunications on its e-mail and
telecommunications systems.
----------------------------------------------------------------
The information transmitted is intended only for the person or entity to
which it is addressed and may contain confidential and/or privileged
material.  Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by persons or
entities other than the intended recipient is prohibited.   If you received
this in error, please contact the sender and delete the material from any
computer.




More information about the thelist mailing list