Lazarus web site done with Lazarus ?

[Fred vS]

Hello everybody. 

Now that Lazarus is adult and has his own web components and everything to create from a to Z a web server, i suggest to translate the Lazarus Web site from Simple Machines into something totally FPC-Lazarus.

Brooks library has everything to do it and it could be a nice example to show the web-Lazarus page done in pure Lazarus-FPC.

PS : I do not have anything against the actual Lazarus site, but it is strange to use Simple Machines who is not pure Pascal, while Lazarus has everything to do it...  . 

[JuhaManninen]

What exactly would you do with Brooks? The forum I guess. The main page is mostly static, or at least it should be mostly static.

Your post had a good timing because I was just discussing about a sub-project to improve the Lazarus main web page. I am going to ask somebody to administer it. I was not thinking of changing the forum platform though. The forum works ok and Brooks is not a forum SW (or is it, I don't know for sure?).

The main page should be more friendly to new visitors. For example currently it does not answer the question "What is this about?" etc.
I am planning to write more about it.
IMO the development process of this open source project works reasonably well, except for the web page design. It needs fix-up, its original authors are busy, and nobody else can change it by sending patches or any other way.
I want to change that. As you say Lazarus is adult and thus the web page could start attracting people as well.

Regards,
Juha

[Fred vS]

What exactly would you do with Brooks? The forum I guess.

Of course, the Lazarus forum is the soul of the Lazarus project.

Lazarus merits a soul done by himself and from himself... 

[Fred vS]

First suggestion for Web site :

-1 Colour.

Could it be possible to imagine something else that the always blue-Windows like colour ?

[JuhaManninen]

First suggestion for Web site :

You can make suggestions to the coming site admin. I am not planning to do it myself. I am not good in web design. How about you?
If yes, then you are a candidate.
There will be a copy of the current VM to play with. There is no strict dead-line because the current page will be there until the new one is ready.
However, the person must have commitment for the task. It also means that all plans must be realistic. For example making a new forum platform is a big task.
It is easy to suggest but actually doing it is a different matter.

Juha

[Fred vS]

How about you?
If yes, then you are a candidate.

Hum, the only thing i can do is a site with GoogleSite   

Maybe Silvio from Brooks knows it better 

The actual Lazarus Forum is working very well, but if the team has a plan to modify the web pages, i suggest to use some Lazarus component to do it. 

Simple Machines is a very good product but a Lazarus one should be better...

[JuhaManninen]

The actual Lazarus Forum is working very well, but if the team has a plan to modify the web pages, i suggest to use some Lazarus component to do it. 

No, "the team" is not going to modify it. Instead I am looking for a new person for the task.
It is so easy to suggest many things when you are not planning to do it yourself.

Juha

[felipemdc]

Could it be possible to imagine something else that the always blue-Windows like colour ?

I like the current color. It is pleasant and has a good contrast.

[jwdietrich]

The main page should be more friendly to new visitors. For example currently it does not answer the question "What is this about?" etc.
I am planning to write more about it.

I fully support your intention to improve the website.

Good (in my opinion) examples for websites of open-source projects are:

• http://www.ubuntu.com
• http://www.suse.com
• http://www.postgresql.org

To return to the point regarding the forum: You are right, a Pascal-based solution would be preferable. However, the effort to transfer more than 8 000 user accounts with accompanying settings and more than 100 000 posts in more than 10 000 threads would be enormous. This could only be done if the new, FPC-based forum system would be in some way "compatible" with the SMF-based infrastructure. I am not sure if there are technical and/or legal obstacles, however.

[JuhaManninen]

Good (in my opinion) examples for websites of open-source projects are:

• http://www.ubuntu.com
• http://www.suse.com
• http://www.postgresql.org

They are OK but not exceptionally good. Suse and PostgreSQL are already too messy for a casual visitor.
On the other hand, those are well-known projects and many people already know what they are when they visit the pages.

Lazarus project is less fortunate. It is almost unknown. Hence the main purpose of the project main page is to attract casual visitors who don't know anything about this project.

First the visitor wants to know what this page is about. There must a clear, short explanation: "Lazarus is ...".

Second, if the visitor is still interested in this newly found SW, he obviously wants to try it. So, there must be an easy and obvious way to install the SW.

Any other information can be behind links. For the first time visitor they have secondary importance.
Experienced users already know which sub-page they want and the main page is not important for them.
For example I don't use Lazarus main page. I go directly to forum, bug-tracker and Wiki.

I have used 2 sites as good examples that fulfill those requirements. I copy them here, too.

1. Our competition http://monodevelop.com/
There is first a big download button and then a clear explanation:
"MonoDevelop is a ..."
Almost perfect.

2. http://www.teamviewer.com/
Commercial SW but they have a free version to attract people.
The download page even recognizes the client OS using JavaScript and offers the right download by default. Cool.

To return to the point regarding the forum: You are right, a Pascal-based solution would be preferable. ...

I don't know why people suggest this kind of things when there is even no forum SW made with Pascal.
The right way would be to start such a project, maybe build it on top of Brooks, and when it is ready after few years, then discuss about using it for Lazarus forum.


Regards,
Juha

[marcov]

I like the monodevelop site.

The teamviewer site is too much a "sell" layout for me.

IMHO have a simple frontpage site like monodevelop, and then move the whole current site under a "development site" option would be ok. That allows both to experiment with the new site without sacrificing anything, with a possible bet the company change that could make the site and forum defunct for weeks.

[JuhaManninen]

IMHO have a simple frontpage site like monodevelop, and then move the whole current site under a "development site" option would be ok. That allows both to experiment with the new site without sacrificing anything, with a possible bet the company change that could make the site and forum defunct for weeks.

The plan is to copy the whole VM and play with it. It does not break the existing pages.
Helmut at FirmOS has promised to copy the VM.

Juha

[jwdietrich]

On the other hand, those are well-known projects and many people already know what they are when they visit the pages.

You are right. Despite their popularity, however, both Ubuntu and SUSE pages explain what it is about directly on the front page.

[snorkel]

The current forum works great and wouldn't it take a long time to implement something as feature rich with the Brook framework?
Plus the embarcadero forums are not powered by Dephi, but rather Java if they don't eat their own dog food for the forum, why should Lazarus?

[BigChimp]

If you ask me, writing forum software in FPC just because we're a programming language site is ridiculous not such a good idea.

Why?
- You're reinventing the wheel: good forum software is widely available (and what's wrong with SMF anyway)
- You get to fix all security and stability issues yourself. Better have a good well-secured backup.
- Time spent on writing forum software might be better spent on other projects (improving Lazarus, the compiler, Brook, Lazreport.. etc) or, to keep it to the website: improving documentation search, cleaning up old wiki pages etc.
- Why stop at the forum software? What about the web server and the underlying OS? Where do you draw the line?

Given the above, I'd think the current project owners/website maintainers/core developers would like to see a lot of assurances before agreeing to implement this.

I suppose that's why the other language sites don't use their own forum software either.

Of course, if somebody was going to write forum software in FPC anyay, using it once it proves to be stable and secure would be less of an issue.
Also, I realize this is all hobby/volunteer work, so I'm only suggesting my personal thoughts.