|
Post by Carlos Phelps on May 27, 2009 21:05:24 GMT -5
If we where to build a new Greymatter from the ground up, call it Greymatter 2.0, what features would everyone want to: - Features to lose
- Features to keep
- Features to add
|
|
|
Post by marlinnut on Jun 15, 2009 23:35:16 GMT -5
Two things:
- a database back end
- normal levels of software support.
I'm totally appreciative of everything you and others have done to keep this app alive for so long, but even the "new" Greymatter is too long in the tooth. But more than that, the lack of a database means waaaay too many discrete files that require backup - what a horrific pain in the ass. It's why I'm in the process of hacking a WordPress template so it'll look like my existing Greymatter installation. I'll be sad to go, but worry that if I don't go now, I won't be able to convert to WP down the line as GM becomes less relevant to those who write the converters.
|
|
|
Post by gambatte64 on Jul 2, 2009 6:24:42 GMT -5
I'm happy with GM as it is (especially menues).
Perhaps the code could be made more secure. I'm a Perl dabbler, so I'm not able to critizise the programming profoundly. Possibly it could stick more with the guidelines layed down by Perl - Best Practices by Damian Conway.
What I certainly don't want is a database. I prefer text files. To keep these files small may be a template engine (Template Toolkit?) could be used (maybe it doesn't fit in with the design of GM, though).
Alfred.
|
|
|
Post by Carlos Phelps on Jul 2, 2009 21:46:26 GMT -5
gambatte64, Making any program more secure should be on the top of any features list. I am glad that you listed Damian Conway's book. I have a copy sitting on my desk and would suggest that anyone wanting to write better Perl program, read it. Should Greymatter have a database or not? The simple fact of the matter is that any program that stores data, files, or info does have a database, but should Greymatter have a formal database like MySQL or loose text files like it has now? I thinks it would be nice if the end user had a choice: - Database types
- Template Engines
- Other modules
Any other suggestions?
|
|
|
Post by stryker on Jul 5, 2009 4:16:06 GMT -5
Hi, and apologies in advance for all the text to follow...
My suggestion might not be so sexy from a technical perspective, but I think if you're considering greymatter's future, it needs to have a growing and healthy userbase. This both assures its relevance (and return on the investment of time and effort from the dedicated few developers we rely on so much) and also reduces the effort required from these developers to also support it. A healthy community becomes to a greater extent self-supporting (and self-marketing).
Which means it needs to attract new users. Which means a focus on simplifying installation and upgrade, and the ability to integrate with or add content from other file types and apps. eg the ability to create a xhtml page in word for example and have it imported as a post rather than someone having to log on and learn a new and unfamiliar greymatter UI. I'm not necessarily suggesting a toolbar with "post this to greymatter" but that level of integration is largely expected and wouldn't be out of place if at all possible.
Scanning through the posts in the GM forums (fora?) will probably give you the biggest ideas of where the improvements should be. If installation or upgrades were able to be automated to a larger extent, the support requirements for that aspect would drop off, and the first impression from a new user would be superior therefore encouraging new users via word-of-mouth.
Integration with RSS, and even facebook/twitter should also be considered as without them the product is less relevant in a modern web world.
The ability to customise greymatter and then export and publish a 'theme' has been explored in the past and I think that would be a great idea. To have a community developing 'skins' effectively keeps the product renewing without any developer hours expended.
Finally my thoughts are whether the back-end is database driven or flat file, or lots of files it's the simplicity and functionality of the front end that will count. So long as there is an ability to backup/restore/import/export to other contemporary platforms I don't have a vote for any particular implimentation.
Thanks for reassuring me there are thoughts being put toward the future of the platform!
Cheers ! Geoff
|
|
|
Post by scottlindberg on Aug 9, 2009 12:39:12 GMT -5
Here's one tiny thing I'd like to see changed:
Right now the default "maxlength" for the author's home page field in the Edit Authors panels are much too short -- 30 characters. For future builds, it'd be nice to see longer URLs allowed. For my own installations, I modded gm.cgi to allow 300 characters by changing maxlength="30" to maxlength="300" on lines 963 and 1229 (in GreyMatter v1.8.2).
|
|
|
Post by scottlindberg on Aug 9, 2009 22:54:57 GMT -5
Another small thing that would be helpful: the ability to rebuild connected files one at a time. I use lots of connected files, and they really slow things down when posting a new entry and everything has to rebuild. If there was an option to choose (via a dropdown menu? a list?) which connected files should be rebuilt it'd make things much quicker.
For me at least.
|
|
|
Post by scottlindberg on Aug 10, 2009 0:21:42 GMT -5
One more thing ... it'd be great to have access to a built-in template that would let you have entry list pages for each of the site authors. Right now I do this via connected files, but it'd be a great added feature.
I'll quit now. I promise.
|
|
|
Post by Carlos Phelps on Aug 12, 2009 23:16:15 GMT -5
Scott,
These are all good ideas.
|
|
|
Post by petefinnigan on Aug 27, 2009 5:05:30 GMT -5
Carlos,
Hi, I have not been active here for quite some time as I am very busy with my company and life in general but i do still use GM on three blogs and would like to see it remain. I started work on creating a built in RSS, RDF and Atom feeds code some time ago but lost interest mainly because I had doubts about doing it and whether any involved effort (there was quite a bit of code written and effort still to do) would be better spent on a new improved GM.
My view at a high level would be to keep the formats of the files, templates, pages etc and move the storage to mysql and possibly also move the code base to php as it just seems easier to me to write in php.
let me know if you need any help with new renewed efforts, I wil try and make some time avaolable
cheers
Pete
|
|
|
Post by petefinnigan on Aug 27, 2009 5:59:30 GMT -5
|
|
|
Post by petefinnigan on Aug 27, 2009 6:02:20 GMT -5
Sorry, three posts in one day. I also just remembeed that someone started a GM 2.0 a lot of years ago on the old forums. I think it was Linear but I may be wrong. Is it worth trying to find him and ask what he did/ can we use it?
cheers
Pete
|
|
|
Post by Carlos Phelps on Sept 8, 2009 22:54:38 GMT -5
Pete, I like MySql for a database, too. But perhaps some people would not have access to MySql or have another database that they would like better. Have you ever look at the Perl module DBI www.perl.com/pub/a/1999/10/DBI.html? It would allow us to write one common Database Interface code to be with whatever Database Driver that the user picks.
|
|