Gossamer Forum
Home : Gossamer Threads Inc. : Discussion :

Links SQL v2.0beta

Quote Reply
Links SQL v2.0beta
Hello Alex!

Whenever this version is going to be out, I would like to be the beta tester.

Basically, I do hope that by this time you would have done the Build_Update mod. Since ONE Year after you installed the script on my server, I am not able to use.

Quote Reply
Re: Links SQL v2.0beta In reply to
Hmmm, seems Alex would like to have all immediately instead bringing up one product after the next one ...
i could imagine which work is needed to programm Links and Mail and DBman and the spider and mybe 2 Mio. other scripts.
I wish his tactik would be the right one; maybe a lot of people would change to another script, cause they are very unhappy about the support ...

I dont understand this behaviour.

Quote Reply
Re: Links SQL v2.0beta In reply to
I do understand his motives, and I agree with them.

I don't think he should have tried to develop several new products at the same time he was working on the central logic, but sometimes it's the development of these other projects that show the weaknesses -- or needed features -- of the core logic.

If you notice, the current version uses some core logic to define many complicated stepts, and to put them into a single, simple function call.

As we all know, not all of the features worked all the time. But it was a 1.0 release, and the big fixes were made for the 1.11 release, and it _is_ stable. It's just not what everyone wants.

What is the links program? It's a means of putting data into an SQL database, and getting data out of that database.

The more, and more robust and flexible, the utilities are to access and manipulate, and output that data, the better the programs, and the easier it is to write those programs.

Rather than hacking the source code in 20 different places, if you can change one central core logic area, you have more stable and maintainable code.

If all the "application" code runs on this core logic, then as technology and algorithms change, you can update, upgrade, and tweak this core logic without affecting the application code -- the same way you can tweak, change, and manipulate the templates to change the look of your site, and not have to change any source code.

Those of you who used the site_html.pl file, rather than the templates with Links 2.0 know what I'm talking about.

Anyway, Alex is guilty of the same thing we all have been at times -- believing we can deliver more in shorter time than reality will allow. Computers are not perfect, and sometimes no matter how hard and how many hours you put into it, you can't get the code working. It's not like laying out a newspaper or typing story. The computer itself (and the network, Mysql, perl, etc) all add layers of complexity to the 'idea'.

I'm hoping for a central core of utilities that can be added to, so that we can create plug-ins to do extra work, add features, change features, etc without having to re-do our entire site. A way of adding in any number of features -- forum,mail, chat, auctions, news, -- anything -- simply by tapping into existing modules, or writing new ones that work with them.

Big dreams... but reusability of source code is an idea I bought into 20 years ago, and I'm more convinced of it now than ever before.

Objects and modules have expanded that. These are exciting times.

FWIW -- if you can find a betterprogram, or support, go for it.

I like what I've seen here, and I'm willing to keep my fingers crossed for the next week or so to see what happens.

FAQ: http://www.postcards.com/FAQ/LinkSQL/

Quote Reply
Re: Links SQL v2.0beta In reply to
Having just done a major change to one central area of my site, the knock-on effect of updating URL's into this area, has taken me nearly a month to complete.

It took so long because of the way I'd developed the site/s in the first place, using exact filenames, instead of relative/directory names, and part of why I've taken nearly a month to complete it is because I've built-in a lot of new directories that will allow me to change things in the future without having to reflect it across nearly every single page of my various sites.

I knew what I wanted to do, I knew it was going to be a large-scale change, but it HAS taken me much longer than I had originally estimated and HAS cost me a lot of time, effort, and most likely visitors.

I'm really pleased with what I've achieved, and a bit disappointed that I couldn't have got it done quicker, but the end result is a new site structure that should last me for at least a few years and I know now that if I need to make changes, they'll be a lot easier than previously.

I think this just relates to how Alex is currently developing his software, he know's what he wants to do, its taking him longer than expected, but the end result should be well worth the wait. Smile

All the best

Quote Reply
Re: Links SQL v2.0beta In reply to
You are both truly right with the things you say; for me is the problem now what to do.
A lot of people wait for me to start a whole relaunch of our main site; all the steps behind my work, marketing, flyers, clients and new clients finding ... everybody waits for me.
If i realize the situation right Alex would need another four weeks for a beta, and then shurely another eight weeks till a stable version (BTW: he wants to go on journey :-(

So what should i do now?
Building the thing with 1.11, doing all the changes and cant simply upgrade cause there would be so many mods inside ???

But it seems i must go this way ...

Quote Reply
Re: Links SQL v2.0beta In reply to
The only real solution is to wait. Hopefully there will be a release in a week or so, and there will be a clue to the logic. That is even more important to see at this point than to have fully working code. So much is changing, and I'm only guessing on how (based on some old conversations). I'm sure things are going to look a lot different than I'm imagining.

It makes no sense to start site development in 1.11 with the next release so close. Studying the code, updating the templates to the most "modular" form possible, and things like that is about the only useful thing to do.

This is the worst place to be. Because the next version of code is going to be so different, you can't really work on it the way you could from 1.0 to 1.02, to 1.10 and 1.11. This will be a 3.0 release, if what Alex says about the new core utilities is true.

I'm working on the catalog system, a bit here and there, doing things that won't change much from one release to the next -- database inter relationships mostly.

So much to do, and so many missing pieces.

FAQ: http://www.postcards.com/FAQ/LinkSQL/