Hi Giles,
Thanks for your prompt and useful response.
In terms of evading the RTFM issues, I would recommend not only a "tips and tricks" section but also something like a "quick start" guide. I assume ht://Dig has found much limelight amongst newbies due to it's mention in the Red Hat manuals. Red Hat customers are looking for free, powerful, out-of-the-box solutions - preferably in rpm format. So when you write a package that simultaneously holds as much promise for functionality and complexity behnind-the-scenes as a search engine, you're bound to see enthusiasm degrade into nagging. I would encourage, as much as you can, the "out-of-the-box" mentality when approaching documentation. Unfortunately, few people seem to care what string concordance algorithm you are using - they want a CGI that they can plunk down in a matter of minutes to snazz up their site. And of course, for the more experienced computer scientist it is good to include the more esoteric details.
I wish you greater success, and fewer redundant questions. Thanks for all your help and care.
Best,
Robert
Robert Peake
MSIA Information Systems
[EMAIL PROTECTED]
http://www.msia.org
| Gilles Detillieux <[EMAIL PROTECTED]>
03/29/2001 11:22 AM
|
To: [EMAIL PROTECTED] cc: [EMAIL PROTECTED] (Gilles Detillieux), [EMAIL PROTECTED] (ht://Dig mailing list) Subject: Re: [htdig] Re: exclude by tag type |
According to Robert Peake ([EMAIL PROTECTED]):
> The email address is listed directly on the web site, and since this was
> not a "bug" per se I didn't enter it in the bug tracking system. If, in
D'oh, I missed the mention of this on the TODO.html page yesterday when
I was looking for it. I need to trust my eyballs less, and grep more.
The note on the TODO list is definitely wrong, and I'll correct it.
Thanks for the heads-up.
The bug tracking system does indeed have a category for feature requests
(which I'll mention when I fix the page), so either it or the htdig-dev
list are usually the best places for feature requests.
> fact, this suggestion is already possible using noindex_start and
> noindex_end, you might post this with your comments tot the mailing list.
It seems that something like...
noindex_start: <frame
noindex_end: >
should do what you're asking, as long as you're not already using these
attributes for something else. A pending feature request is to expand
these attributes to use string lists, so you can have multiple sets of
strings to remove. If this doesn't work, let us know.
> Also, you might want to conisder publishing a "ht://Dig Tips and Tricks"
> section that offers configuration tips for various search engine schemes.
We're currently discussing various ways to make the documentation more
newbie-friendly, so any contributed tips and tricks would be most
welcome. We could really use a write-up on setting up multiple databases
for a virtual hosting environment, as we've had a few requests for this.
We could also use more search configuration examples, especially for
radical customization (but still sticking to the standard htsearch CGI).
> The system seems very powerful, and I have a feeling I'm not the only user
> that is harnessing a mere fraction of it's capabilities.
I don't doubt that. It tends to be true in general of most software
packages. However, some documentation improvements would likely help
users find out more about what they can do. Of course, it'll only help
if users actually read it. I've found that almost half the questions
on this list are already answered, fairly directly, in the FAQ or in
the documentation. What is not clear is whether those asking the
questions have spent any time at all looking for answers before asking.
In some cases it's quite obvious, but I suspect that in others they've
taken a look but had trouble finding what they're looking for.
--
Gilles R. Detillieux E-mail: <[EMAIL PROTECTED]>
Spinal Cord Research Centre WWW: http://www.scrc.umanitoba.ca/~grdetil
Dept. Physiology, U. of Manitoba Phone: (204)789-3766
Winnipeg, MB R3E 3J7 (Canada) Fax: (204)789-3930
