Excited to Join New Forum – Proposal for a USA-Based Collaborative Wiki for Freely Available FIGU Information

[james]

Patrick, what you did with AI and the contact reports I find to be interesting and good quality. I've read 6 so far and found no obvious errors. Only a little thing about the Destroyer still being a threat but it no longer is thanks to the Plejaren's actions.

If you could create a bot to do the posting then that would help the mission significantly and not consume anyone's time a great deal, once its gotten started. How hard would it be? The only problem I see is funding of the premium tier in order to get maximum coverage. Could you easily afford it? If not then we could probably generate the funding here.

These AI posts are varied posts and I don't think x.com algorithms like monotony, people posting the same thing over and over, and so I think that your AI generated posts would also be positively received by the x.com technology and staff moderators.

Many thanks,
James

[Arthur]

Recently i wanted to develop an application from contacts report.I did it. I can download the contact from... to... Despite the fact that the format is initially HTML, I was able to extract the text with the headers, but there is still a lot of useless information.
Maybe a JSON format from HTTP request for the contact with headers (date, etc...) according to selection coded in the url (language, all contacts, from... to..., updated since a date specified) will be more appropriate.
It is simply an idea for developing applications that can read contacts.

[KelseyBryantAnthis]

Arthur,
Nice work getting that extraction working from the HTML! If the issue is the “noise” in the data, you’ve basically got two main options:

1. Refine your HTML parsing
- Use a parser (e.g., BeautifulSoup in Python) to grab only the specific tags/sections containing the content you want.
- Strip out styling, navigation, and unrelated metadata at the extraction stage.
- Keeps compatibility with the current HTML export.

2. Create or request a cleaner feed
- If the backend can be queried, propose adding an API endpoint returning JSON with only the needed fields (date, language, contact number, content, etc.).
- Parameters could include: ?from= / ?to= for date ranges
- ?lang= for language filter
- ?updatedSince= for incremental fetches
- If a public API isn’t possible, a simplified export template might be.

Tip:
Mock up what your ideal JSON output looks like based on a couple of real contacts — that makes it far easier for others to see the potential and decide if it’s worth implementing.

And to push the idea further…
Modern AI tools can help prototype and iterate quickly. For example, you could:
- Feed a few sample exports into a model like Claude (Anthropic), Microsoft Copilot, or ChatGPT to generate draft parsers, regex patterns, or API specs.
- Use AI to clean and restructure data automatically, flagging inconsistencies or missing fields.
- Let the AI propose naming conventions and schema designs for your JSON so downstream apps stay consistent.

This combination — a cleaner data source plus AI‑assisted iteration — could make your “contact report reader” idea a lot more powerful, and far quicker to evolve into something others can build on.

[Arthur]

1. Refine your HTML parsing

No, because it is not formatted. I used a reference at the beginning and at the end to delimit the text, but it is not present in all contacts. ("Next Contact Report" for example).
The way i did it is not the way it should be done.

A HTTP request can be made to obtain the JSON data, but there is another method :

Knowing the data are static (archive in fact), a simple way to do it will be to setup an old good one public FTP server. (port 21)

On server side :

- Create a batch who read the database, write the json file, (1 per language), compress it (xz compress better than zip) and put it in the FTP server.
- There must be some kind of nomenclature in the file name for the contact number and language. Contact_1_e.json.xz , Contact_915_g.json.xz, Contact_latest_e.json.xz (the same as Contact_915_e.json.xz as of today) etc ...
- A read only server.(no put, create dir etc...)

On client side :

- Connect to the public FTP server (with anonymous, guest; as user, pass).
- List the directory to see if there is new modifications.
- Get the file, decompress it, parse the json data.

Today with the 915th contact that makes 1830 files.
I don't know the average size of a contact but with 100k , 40k compressed gives less than 100 MB.

Just an idea to open up contact reading.
This way, anyone can read contacts at will, on any platform.

[Pureharmony]

Hi, I hope this board turns into the next Figu Forum and stays around as long as that did.

Hi Norm!!

I remember you from the Figu forum way back.

-Pureharmony

[Hush]

Hello James,

I appreciate your enthusiasm for contributing to the FIGU mission in some way however I'm not sure that we need yet another collaboration website.

Surely this forum is a sufficient type of collaboration system where we can discuss ideas on how to contribute to the FIGU mission further. Does it lack a feature you have in mind?

This forum has absolutely zero chance of facilitating productive, engaging, positive collaborative discussion to support the FIGU mission as long as it permits endless streams of garbage advertising unrelated to the forum theme.

Forum_Mess.gif (96.61 KiB) Viewed 185 times

What makes any forum successful is implementing moderators who enforce a basic policy whereby posts are regularly screened to align with the forum theme and subject matter otherwise said content is removed promptly which DOES NOT exist here. Forum moderators is principally why the origin FIGU forum was a success and dare say, necessary for a forum to prosper, attract participants be engaging and inviting. As it is, this forum lacks such integrity and prone to becoming a complete MESS because there are absolutely no moderators keeping the house clean of all this advertising garbage.

[james]

Yes I agree. Would you like to become a moderator?

[james]

I've increased the difficulty of user account registration.

[Hush]

Hello James,

Your revised account registration policy may be just the ticket to reduce or eliminate these annoying advertising posts - thank you. For now, perhaps just wait and see if its effective however if the problem persists, am willing to help out if only to keep the forum house clean of unsolicited non-FIGU related advertisements.

[james]

Agreed