Gossamer Forum
Home : Products : Gossamer Links : Development, Plugins and Globals :

Ratings Review Group Project - Pre announcement

(Page 1 of 2)
> >
Quote Reply
Ratings Review Group Project - Pre announcement
Because people can't wait .... :)

The Ratings/Review mod is a really complicated modification, and everyone has a different idea about what it should do, and how it should look. There is no way that one programmer, or even two, can do it all.

This is a prime candidate for a "group project".

Alex has agreed, and has, or will, set up some areas for working on this. I will most certainly be a part of the programming group, and I'm sure Alex will lend his support and experience, but we really need a bunch of people willing to take on various aspects of the program. Consider this an "open source" product add-on/modification to a commercial product. All the rules for group development need to apply!

To start out with, we will need everyone to post their ideas, suggestions, needs, wants, wish lists, essential features and 'like to have' features as much as possible BEFORE planning starts. If you want it to make your coffee in the morning and walk your dog, now is the time to suggest it.

Development will require several groups of people working together, and they are not all programmers. We need people who can take all the comments and suggestions posted in the forum, and create a proposal of what the program really needs to do, what most people want it to do, and what special needs it needs to consider. As new suggestions are posted, those have to be incorporated, so that there is always a "reference doc" for the developers go to, without having to read through the forum.

This "front end" group is an ESSENTIAL part of the process, that filters out the stuff people post, and turns it into a form that the programmers can work with. This doesn't mean the programmers won't read the forums, but it's much better if the programmers stay out of the "consensus" group, at least at first. Now is not the time for trying to decide what can and cannot be done, but rather what everyone WANTS done -- realistic or not.

This "front end" group will have the key job of dealing with the daily posts, taking the new releases and presenting it to the people, and then feeding back the problems. They are the "interface" between the programming group and the user community. These people are very important. They need to know how to read and write, how to use links at least a little, and they have to be able to keep their cool under fire. Their purpose is to prevent the programmers from getting side tracked in all the inevitable tangential "problems" that come up.

Group 2: The Programmers. These are people willing to work in a group, and either help design, or debug, modules or code segments. During development, modularity helps, but once release time comes around, many modules may be combined for performance reasons. This group needs to get along with each other, and be willing to work together to create the interface and rules under which the project develops.

Group 3: The HTML/Graphics/Template group. This group will be responsible
for creating the templates and HTML, as well as default sets of graphics to
be used with the program. This is an important group, since this is what the
"user" will see. This group needs to work closely with the first group, and
the users, to respond to what they need and want. They will have _limited_
need to contact the programmers -- the interface should be well enough defined
that the only communication is if the Group 3 people feel something needs to
be added or changed to make the program better, or if there are of course bugs.

Group 4: The users. You know what you do! You are to be nice to the people
in Groups 1-3 since they are doing this for you, out of no monetary or other
gain.

Hopefully, this is a brief intro to what is needed, and we'll pick up at least a half dozen people who want to be in Group 1 to start dealing with the wish lists for the next week or two.

Early next week, there should be a forum for discussing this, and a means for us to distribute the code.

For now, you can post your comments here, to this thread, but mostly, this thread should be a show of hands with a willingness to participate and what you can or would like to do for the project.

Don't be misled. This is a _BIG_ project. It has to be flexible, and handle a lot of variations. The underlying goal of the project is to create a bunch of flexible routines that can be used to allow people to easily customize how they want their reviews to look and to function. No "hacks". Real, workable, and well designed code, logic flow, and graphical interfaces.

The other benefit of this project is that it will show potential developers and 3rd party developers what it takes to build a plug in! So, everyone should benefit from this.


PUGDOGŪ Enterprises, Inc.
FAQ:http://LinkSQL.com/FAQ
Forum:http://LinkSQL.com/forum
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
Well i can certainly help. I think i am quite good at part 1, understand the idea of programming and links but not actually good at doing the programming itself. I can of course do 4 and some of 3 as well.

Some things/features i can think of at the moment:

2) User settings / validation
- require registered user/guest
- require validation/no validation
- if require validation, ability to over ride on specific users.
- this override on user basis, possibility for automatic after 10 reviews.
- possibility of seperate usergroup to registered being reviewer. If you register to add links you need to click another button to register as a reviewer as well? seperate the two? perhaps not...
- people reviewing are not going to be as "dedicated" to register as those adding their own links so include the registration form in the first review.
- Rating/Reviewing seperate. So that a person can rate a link as a guest but cannot review until registered. Users include rating off site (cgi-resources.com style) but require registration to add review. System to let someone rate then ask them to add review (use the signup form now)... using a session/hidden form so that if they do complete the review the previous rating is deleted and added to the review. (sorry not well explained)

2) Editor's new options

- editors can validate reviews in their sections
- editors can modify reviews
- editors can delete reviews
- editors can add reviews
- editors can give user permissions to add reviews

3) Fields

- unlimited custom fields. These are designed in the plugin system
- Field types all specified there as well
- Review form is built dynamically based on these fields
- Some fields required/others now required.

4) Ratings

- seperate from review, i would think have to two tables like:

Rating ID | Link Id | Rating

and as part of review have field Rating ID

this is so:

- (see previous section about allowing rating for guests but not for reviews..)
- easier to have multiple ratings...

- Multiple ratings - ie Design/ Features / etc.
- Allow to set logic as a rating so for example rating field 4 = avg(rating field 1, rating field2) etc.
(so you can create a dynamic average for example for sorting later)

5) Link options

- need the information avaliable to link.html so that number of reviews/ratings etc can be used in the link.html template.

6) Display of ratings:

i) Graphical display of ratings, much like widgetz mod
ii) spanned display list of reviews. Admin setting of number of reviews per page.
iii) admin setting of review form at bottom or require a new page for this.
iv) use of review page like a detailed page (see special reviews for more info on this)

7) "Special Reviews"

- special reviews but admin/editors
- special status to review, changes style of review be the review in bold or have a special image. seperate template then.
- might be used as a detailed page style with the rest of the reviews listed below.


Well that is all that I can think of at the moment. I think all those options would give quite a lot of flexibility but I am sure some options have been missed!

http://www.ASciFi.com/ - The Science Fiction Portal
Quote Reply
Re: Ratings Review Group Project - Pre announcemen In reply to
i am currenly writing one for a client that includes stuff like:
only show 4 reviews on the detailed page, then show 10 reviews on each page there after.
users can only review link once.
Overall star rating.
User must be registered to review, either by having a link in the database or by registering.
a reviewers review count is counted, haha....so it could be possible to put a star or whatever next to the users name if they reviewed 10 links or whatever.
validation should be an option for sure of the reviews.
either an admin or an editor of the categorie get the email.
being able to edit the review or delete the review for those hoodlums out there.
bleh bleh bleh...



Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
Hi Pugdog,

I have the next month free so I am happy to do the templates and let you big boys get on with the coding.

Paul Wilson.
http://www.wiredon.net/gt/
http://www.perlmad.com/
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
This sounds like a great idea. Since I'm just a beginner at Perl, I can only help out in all groups except for 2. However, I won't be as free as PaulWilson though.
I have a couple more suggestions to add to Padders list:
- ability for reviewers to edit their reviews at a later date
- ability for readers to send an email to the reviewer without knowing the reviewers email (ie. via a form, checkout my site at Scutwork.com).
- or even better, have the reader post the questions online and have a notification email sent to the reviewer. This way, the reviewer will not receive the same questions again and again.
- An option for reviewers to allow readers to post questions to them.
- a separate page for the review form.

Adrian

Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
I have also ssome ideas:
1. User modify options, each user can modify their username & password, and eventually some extra options depending on the fields you have in your user table (hobbies, ICQ, occupation)

2, internal message system for registered users, editors
registered users/editors can send, read, reply, delete, modify messages, like the system in message boards, but this is a cool option for Links-SQL users who don't want to purchase a message board

3. editor validation:
registered users van sign up for editor for a category with their messages, why theu should be an editor in that category.
after signup, the administrator can validate, delete, modify the editor and sending email.

4. MyLinks, each registered user can create it's own bookmarks with sites in the Links-SQL database,
each link became a button, or checkbox, after user registration, so that user can add/delete/modify it easy in it's bookmark section
------------------------------------------------------------
I have not the experience in perl to build the plugin, but I can redesign the HTML codes & templates for the plugin, that makes it easy for users to get the plugin, and the rest of their admin in their site layout with only changing a stylesheet.

Good Luck.

Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
Hi Pugdog, et. al.:

It seems to me that reviews seem to be what really should be looked at here... a ratings system already exists. If the ratings system needs to be made over or enhanced, maybe that should be a seperate project, seperate from reviews. As far as I am concerned (at least) ratings as they sit are fine...

I wanted to second (and third) some of the above suggestions:

1) There should be a method to allow only registered users or anyone to post.

2) There should be a method of editing your own posts.

3) User profiles available online.

4) Perhaps a "profane Language" filter. I myself am not easily offended, but a LOT of kids visit my site.

At the risk of offending Scream, it might be good to use wwwthreads as a basic outline- not so much for code, but for where to head. (Scream- thak this as a compliment!) I think a lot of us that use LinksSQL use wwwthreads, so it may even be worth exploring integrating the two, at least to the extent that wwwthreads would take the forum names from the Links Categorys. I don't know, I am just throwing out ideas...

Anyway, as I remember, this plug-in was supposed to be a "Basic Review" system, and maybe for all concerned, that should be the thought process. A more extensive review system could be added on later with various plug-ins as they are developed.

Anyway, just my two cents.

Oh, and Pugdog, consider me voilenteered in any way you can use me!

dave

Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
i had thought of the idea of using a forum type system for reviews, in my case i use vbulletin which makes this very easy with unlimited categories for this sort of thing. In the end i thought it was a bad idea for a number of reasons:

i) requires another program's systems which will change over time
ii) added resource use of the other program. Extra threads, forums etc...
iii) likely to confuse forum users.
iv) unversatile. Because we rely on the forum use makes it unversatile.

So those are some of the arguments for building the system from start up. I really think we should use the basic model from the old review plugin with a lot of the other features that have been talked about.

I love the idea of a user control panel type thing. As well as the much requested "change email/password etc" (which is probably a seperate plugin) this plugin would add an extra "tab" to this, review management where you could see all your reviews, edit them, change ratings (if allowed) etc...

So perhaps we have two plugins here:

i) a user control panel. This is different in it is a seperate cgi script that requires the hooks so other plugins can use it.
ii) the review plugin.


The plugin that is being done for a client, can this be used as a basis to start with? Pugdog, started on this? an

http://www.ASciFi.com/ - The Science Fiction Portal
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
p.s.

one thing often asked for seems to be an internal messaging system. Although this i think can be done a lot better using private messaging of vbulletin (or wwwthreads if you must) this would then be a plug in using the user control panel type thing. It is not part of a review system though i don't think.

http://www.ASciFi.com/ - The Science Fiction Portal
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
although i don't participate in Links SQL discussions anymore.. I'm hope I can be of help in this project. Ratings/Review were one of the areas of Links that I was most interested in, hence review.cgi v1 and v2 for Links SQL 1.13.

I have the current version of Links SQL but I haven't really looked into it. I'll keep an eye on this thread and help out when a problem arises..

Jerry Su
http://www.jsu07.com
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
pugdog,

What happened to this project? Did it die? I for one would find this very useful - If I knew anything about perl I'd have thrown my energy behind this one. Hope you're still pushing it. Thanks for all the efforts.

Safe swoops
Sangiro
http://www.dropzone.com/
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
This is not directed "to" you, but it's a general update on this, me, and what I can and cannot do!! :)

Well, I got a couple of people who said they were interested, and they backed out for whatever reasons before anything started up.

I just got the first version of the 03.xx multiple file attachment program released, and I need to clean up any bugs, and finish adding in features to that, as people report them.

I'll be co-developing the reviews system with the 06.xx gallery program, since being able to add reviews, and have a forum attached to each link for comments is very important for a sub-release version of this program.

I run several domain name development sites, and I need a way to attach ratings, reviews, and discussions to each name, as well as sample logo/graphics packages for each. I have about 4,500 domain names, some are "brand protecting" not real development names, but I have at least 2,000 sites to develop, set up, and offer as ready-to-run or "start up" businesses.

But, as I said in the early posting, if I do all the work, I can't release the program for free, and the competition is about $500 (or more?) and this program will be better, and will be pure-Links SQL. I know that puts it out of the price range of most people, but without a group effort, I really am not allowed (legally) to release the program below market value, if I develop it alone.

I've released several ideas for starting up, plans for fields, and how the database/data will lay out, but got very little feedback. And no real offer to start working on any modules, subroutines, or anything. I even laid out a 3 or 4 level call for volunteers, where once a few people start working, there will be "grunt work" for people who don't know coding or perl, to help out.

As the project develops, "lite" releases can be made, like what I've done with the Image Gallery 06.xx series I'm working on. The 01/02 logo/graphic mods are base-level code. Upload one image, attach it to a link. The 03xx code allows multiple images/files per link, and more features for managing the files. The 06.xx will contain everything <G> And I might change the numbering of the other modules to allow an 04.xx and 05.xx release that are intermediate to those.

Without a group effort on the Ratings/review system, i don't even know where to draw the lines, and what features will satisfy "most" people at each level of release. You can never satisfy everyone, but what are the "basic" needs that people want. What are the next level, etc.

Even one or two people taking charge and and finding out what these ares levels of needs are, and polling people for features, etc would be a great start, and help in planning design.

Hopefully, in the forum redesign, Alex will add this as a developers forum, as was discussed awhile back. That might help get it started.

As I said, this would be a gold mine to develop as a commercial project, but I don't have the time. It's a secondary need for me, the image management being my primary need. I can help coordinate, and contribute to the project. I can even lead it, but there has to be a real effort of others behind it doing work :) It would also be a great place for showing others how to create plugins, modules, and hook into links. It would be a great learning experience, and a much better hands-on example than the "search log" plug in or whatever that was <G>.

So to all those who have said (mostly privately) I'm doing this for the money, believe me -- I'm not. I could have had the image mods working on my sites already if I didn't have to develop them so others could install them, so they could be flexible, and survive "standard" upgrades. I'd be working on the ratings/review system rather than my image programs, since many, many more people want that than the file and graphics management programs -- and a whole other different product alignment, but I can't. I have obligations to my primary job, and until any "money" from these forums reaches serious levels, I can't move those primary obligations around. This is the main reason I have asked for pre-orders. If people pre-order, and put their money where their mouth is, I can move my obligations around. Right now, the "money" I make off this forum, on average, buys my kids pizza and a movie on fridays, if I don't take out my "overhead" for being here. That's not an amount I can use to justify avoiding my primary obligations and start work on a ratings/review or "other" type project.

I hope this clears this up.

I realize a lot of people want things, and Alex/GT is really stressed with their new releases, and keeping up with the 'core' features in the products. Right now I think I'm the only developer for Links SQL that takes on "external" projects at all -- and I am _really_ only one person, who is also responsible for all the tech support and development for his company's servers as well. I"m a one-man MIS department <G>. Most of the others offering help here, are developing their sites, and offering support or information, but not external modules. It's a _LOT_ of work to create a plug in. What works on your site, may not work on other sites. The Install.pm module for the 03.xx series, is a bear! And there are loads of error checking and special conditions I still have to add in. It's not unusual for the install program to be 5-10x (or more) larger than the program or module it's installing!

Anyway, if Alex/GT sets up a forum for this, I'll certainly contribute. I'll even help lead it. But really, other people need to jump in.


======= disclaimer =======

As always, it's NOT about money, but it is about money (or compensation for time, really).

I have this pretty well planned out. It needs alot of work, and at least 100 concentrated programming hours to release a reasonable, stable "lite" version of a ratings/review system (whatever that is officially described as). Figure 10x that for the pro version, which would be on par with the big boys like Amazon and the others.

If there is really a demand for this, I need people to show it. Remember, put the money where the mouth is.

I will restructure the next month to develop this, if in the next week I get 100 people willing to pre-purchase it. I'll offer these pre-purchase people a $75 price for the lite version, and a $125 price for the pro version (whatever the final pricing will be). If I drop the price on the lite version for any reason, I'll up grade those pre-release orders to the pro version for free (not likely to happen, but it could, and that way those people are protected price wise).

The "lite" version will be released in 4-6 weeks after the project starts, and the pro version will probably be released at the same time the Image Gallery mod is released. If you want the Image Gallery mod, I'll include the ratings/review system in that, for the $225 pre-release price.

But, I need 100, legitimate pre-orders. I can't go by what only "seems" to be the demand here.

Why??? For all the requests for the multiple file attachments, I only 5 orders for the 03.xx version at the moment, and they are from people who I've never heard from before -- or who have probably ever posted here. Ok? I can't afford this sort of response, unless I myself need the program. I need the image gallery program, I don't need the ratings/review program at this point in time (I will later, and I'll write it later). I have much, much bigger plans for the Image Gallery program, so I'm working hard on it. This is my primary obligation.

===== end disclaimer =====

I hope this helps put a real-world perspective on all this.

I "make" enough from my participation here in the forums to buy my kids pizza and a movie once a week. I have primary obligations that cannot be rearranged, or I'll end up in trouble. I can rearrange them ONLY if I can show real, tangible, and in-hand compensation for that time.

And, in closing, for all this "loads of money" I've been accused of making here by some, how many peoples sites did I get up and working this week alone, or fixed up, with _NO_ exchange of anything but a "Thank you" ???? It feels really nice, but I can't pay my bills on that.

Ok?



PUGDOGŪ Enterprises, Inc.
FAQ:http://LinkSQL.com/FAQ
Plugins:http://LinkSQL.com/plugin
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
A Links SQL compatible rating/review system has in fact been developed somewhat along the lines of the Amazon.com model. I'll make an announcement about this before too long. Currently the program is in beta and I'll be looking for some volunteers from this forum to test it out. It's not a Links SQL plugin. Rather, the Links SQL database would be a plugin to the review program which will be offered commercially (currently just over $200). I'll be developing this program continually, and depending on the level of interest expressed by the Links SQL community, I can more closely align the two over time.

I'll post more on this when I think it's time to start seeking beta testers. I have one Links SQL owner attempting to get the program installed, but currently we seem to have run up against a file ownership problem based on the fact that the Links SQL files are owned by nobody while the review scripts "run as username" (because they were uploaded via telnet) according to my beta tester. So when we figure this out I'll be happier about offering it to others for testing.

This review program is not publically available as yet. Requests to receive the beta version will need to be made directly to me. But please don't attempt any requests until I have posted more on this topic. Thanks.

Random Mouse

Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
Sounds interesting!


Installs:http://wiredon.net/gt
FAQ:http://www.perlmad.com

Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
I think we may have talked about this awhile back, and the philosophy of doing it this way, vs making it a plug-in to links limits how links can make use of the program. I think for lack of a better analogy, this review program is a shell around links, and stands alone, referencing the Links record.

The program I've proposed, and outlined, _is_ Links, and works from within links, to do everything. The Link record is still the basic unit, and the reviews are attached to the Links record via LinkID, and to the User via Username. There is full referential integrety between the various tables, and FK relations can be enforced.

There is always room for more than one way to do things, but I _much_ prefer working with the GT engine code, and doing everything from within links, rather than not.

One advantage of doing it within Links, is that the Authenticate module, will, hopefully, allow one connection point between all the GT programs and the 3rd party modules like w3t and vB, etc. Coupled with the FK referential integrity, it's a more integrated and upgradable pathway.

As for the permissions issue, if you are trying to get around writing to the user_cgi directory, or cgi-bin/links tree, the only way to do that is go in as "you" and manually change the permissions. Even with plugins, this has to be done if the directory was created as "you", and the server is running as "nobody".

The only real way around this I've found is with Webmin, which runs as a root daemon server, so has access to pretty much whatever you need to do.


PUGDOGŪ Enterprises, Inc.
FAQ:http://LinkSQL.com/FAQ
Plugins:http://LinkSQL.com/plugin
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
In Reply To:
The only real way around this I've found is with Webmin, which runs as a root daemon server, so has access to pretty much whatever you need to do.
I was recommended against using webmin by some of the *top* gurus at Cobalt. It wrecked our server as it screws with files improperly.

Webmin cost us $150.

Installs:http://wiredon.net/gt
FAQ:http://www.perlmad.com

Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
Well, the "key" here is "Cobalt". Nothing on cobalt is standard.

I've had good luck with Webmin, and many, many ISP front ends to Linux and Solaris are based on Webmin.

With Sun now owning it, maybe it will become more standard, and Sun will offer lower end, plug-in modules that run Solaris or Linux, in a small space, and a small price.



PUGDOGŪ Enterprises, Inc.
FAQ:http://LinkSQL.com/FAQ
Plugins:http://LinkSQL.com/plugin
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
I'm sure the Cobalt/Sun guys are experienced with a range of OS's and if they recommend against it then who I am to argue.

Don't shoot the messenger Smile

Installs:http://wiredon.net/gt
FAQ:http://www.perlmad.com

Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
it is true that if the entire review program was constructed from Links SQL code it would be "better integrated". that goes without saying. however, the larger and more complex the proposed plugin is, the less likely you are to ever see it developed as such. especially if the developer is third party, like me. too much investment is required. this is also the reason that pugdog's review effort has yet to get off the ground. and he's initimately familiar with the Links SQL program. imagine the difficulty faced by those only passingly familiar with Links SQL!

about this referential integrity stuff: unless the Links ID for a given review item changes for some reason, i don't see any real problems with the coupling between the review program and the Links SQL database (how often do people change the IDs? perhaps Links SQL owners could educate me on that point. and i'm only talking about the ID in the Links table, not any other). at worst, i guess someone could write a little plugin to keep the two databases aligned when ID values are altered (me?).

as for authentication, i've written modules to allow cookie facilitated reviewer authentication on these member databases: Ikonboard, PHPNuke, Links SQL, UltimateBB, VBulletin, and WWWThreads. of course, if i was using the Links SQL modules, life would be simpler still. but this is a stand-alone program that has been adapted for Links SQL users simply because some of them asked me to do so. i have given the idea of turning it into a plugin some thought, but that would take a great deal of time. also, if an alternate review plugin is to be developed, then it is probably not worth the effort on my part to pursue a parallel line of attack. better, probably, to maintain two different philosophies and see where each one leads.

i better not say too much about the program until it's ready to be tested, but after it's released Links SQL owners will be able to download it and try it out before deciding whether they want to purchase a license.

Random Mouse

Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
In Reply To:
it is true that if the entire review program was constructed from Links SQL code it would be "better integrated". that goes without saying. however, the larger and more complex the proposed plugin is, the less likely you are to ever see it developed as such.
Not true :) There are some awesomely complex projects being developed by several parties, each because they need it, and a few are generalizing it, hoping to either recoup some of their development costs, or to make it useful to the general community. I'm a "wholly owned subsidiary" shared by two corporations, so I can't work for free on projects that would have a commercial value. Then again, not many people do work for free -- for long. I plan to be around for awhile. I've outlasted many, many others, and plan to keep doing so.

As for referential integrity, the problem comes from several directions. If you have an external program that is managing reviews of a link in the Links system, and the links user decides to delete a link, the referential integrity is lost. You now have an external program pointing to a links record that no longer exists.

Ah, you say. I'll block links managers from deleting links via links, and require they only delete them via my reviews program. (a complicated process if you don't have a plug-in into links to short circuit the delete option, but I'll assume that exists.) Your program deletes the link. But, without FK checking, it now breaks relationships with the CatLinks table, the Users table, and whatever other tables have been added by "legitimate" plug ins that depend on that Links.ID to be there.

In MySQL foreign keys (FK) are advisory only. They have to be enforced by your program code. They are not enforced by the SQL engine. Links and the GT `engine have 'fk' checking via the .def files, and associated code. This is why using the Links/GT engine to access the data, and to create major plugins is so important, and strongly advised.

Besides simplifying the code, it affords all sorts of benefits you havem't thought of yet.

In Reply To:
this is a stand-alone program that has been adapted for Links SQL users simply because some of them asked me to do so. i have given the idea of turning it into a plugin some thought, but that would take a great deal of time. also, if an alternate review plugin is to be developed, then it is probably not worth the effort on my part to pursue a parallel line of attack. better, probably, to maintain two different philosophies and see where each one leads.
Well, my reviews program is being developed, along with my other ones, and is early pre-alpha, but fairly well thought out. It's been designed to plug-in to links, and to adapt to various uses, since Links is used in so many ways. There are only 24 hours in each day, and I can only get so much done during them.... :)

I believe I offered to turn your program into a Links SQL plugin, and port it to use the GT engine, some months ago, but you felt it wasn't in your best economic interests to take me up on it.

Hence, we have at least our two development thrusts, I've known of one other, and I've just become aware of a 4th.



PUGDOGŪ Enterprises, Inc.
FAQ:http://LinkSQL.com/FAQ
Plugins:http://LinkSQL.com/plugin
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
deleting a Links ID won't cause any problem as my program begins by querying the Links Category table, using what it finds on the fly to hook into the Links table for associated IDs. if the ID goes away, so does any reference to a review. as i said, only if the Links ID value is changed after a review has been approved is there likely to be something of a problem (under what circumstances might this occur?).

my philosophy is that GT is the best party to administer its database. if i stick to read only operations i don't have to worry that my own application work seamlessly with every other plugin or the GT code. but to do that, to be able to focus on the Links SQL database and not the managing software behind it, i do have to make an assumption that at least one relationship is a constant. i've assumed that the ID in the Links table doesn't change, but that every other foreign key might.

i wouldn't discount my program just yet. i'm prepared to let the beta testers do that.

Alex was kind enough to give me a copy of Links SQL several months ago so that i could look at the possibility of turning the program into a plugin. unfortunately i encountered some kind of bug that prohibited be from interacting with the program as a general user, although the administrative part worked fine. so after starting to plough through the code looking for the source of the problem, and not finding it, i got a little discouraged and went back to work on the review program. a true review plugin is not out of the question, but it would be a long term project if it did materialize.

it is true that you attempted to obtain the rights to my software. not to make it available free to the Links SQL community, of course. but the per license royalty you offered me wouldn't have covered the cost of a movie, popcorn, and a small soda. for a product which, as you have repeatedly asserted in this forum, could have a market value in the range of $400-$500, my business sense strongly suggested that i reject your proposition.

one thing seems certain though. with 4 projects to choose from, the winner in all this has got to be the Links SQL owner.

Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
In Reply To:
it is true that you attempted to obtain the rights to my software. not to make it available free to the Links SQL community, of course. but the per license royalty you offered me wouldn't have covered the cost of a movie, popcorn, and a small soda. for a product which, as you have repeatedly asserted in this forum, could have a market value in the range of $400-$500, my business sense strongly suggested that i reject your proposition.
The market value is only if the work is put in, as you have pointed out. What I suggested to you, was a way for you to hit the Links community with no effort on your part. It's a small, niche community, and the overall return on a program like this is what? I'm sure you would have been ahead if you continued your basic development, and not had to target this program specifically these past months. How many hours have you put in, and when it's all added up, will you still have as much as a "the cost of a movie, popcorn, and a small soda" from each copy you sell?

Unfortunately, the "profit" from software sales is very small. The money is made on servicing the software -- installs, upgrades, customizations, etc.



PUGDOGŪ Enterprises, Inc.
FAQ:http://LinkSQL.com/FAQ
Plugins:http://LinkSQL.com/plugin
Quote Reply
Re: Ratings Review Group Project - Pre announcement In reply to
i may well be crazy. either way, i'm going for the big soda.

Quote Reply
Re: [RandomMouse] Ratings Review Group Project - Pre announcement In reply to
Cool

I was wondering how is this "review" plugin progressing? Just want to let you guys know that I will be a paying customer. Please let me know if it is ready. I would be very interested.

----------------

waffle

Quote Reply
Re: [waffle] Ratings Review Group Project - Pre announcement In reply to
hi,

i'll probably post information on my review program about the same time that GT puts out Links SQL 2.1.0 with their new review plugin.

actually, mine's not a plugin per se, so it won't work the same as the one GT makes available. i've been spending a bit of time optimizing it which is why i've held off offering it to this point. i'll just have to see whether there's any interest there or not. otherwise i guess i'll have to court the Hyperseek crowd(!)
> >