Wanted: a Libmaster - no pay, but glory

and not very well at that (I apologize - not enough time).

it over. We can bounce ideas around, but the Libmaster will have full control over the library, how it works, change policies, postings, etc. The server will continue to be at reboltech.com, which has good backbone connections, so it's a good spot.

of young kids) who are interested in REBOL, but they find that a lot of the scripts in the lib don't work (because they are examples, not working scripts -- so they should be in an example section, apart from other scripts.)

useful to others, and we need a reliable place to put them.

is required, and there's no pay for the job, but a lot of glory... major historic REBOL glory. ;)

I think a lot of us are in this situation. Rather than have a single librarian, why not have a commentary system whereby anyone can post comments on a script within the library, and suggest fixes etc. Many view scripts broke with successive new releases of view, but they remain in the library. And for new scripts to be included, perhaps one could have a system whereby script submissions have to be vetted by a senior list member before it gets accepted. Otherwise it sits in the pending folder. -- Graham Chiu

Yes good idea. Recent threads on Vanilla/wiki and "browsable system object" suggest how this community library can be engineered and maintained, with final reviewer[s] as you suggest. ./Jason

Zope has 'ZopeLABS': http://www.zopelabs.com/ Digital Photography has http://dpreviews.com <==AWESOME good for content, features navigation and smart fast friendly HTML. Its useful to evaluate various features and criteria. 'Submit' and 'Community' are two important ones...here's my 0.02 for 3 sites I know and use. hth ./Jason =============================================== PYTHON

Great site! - Integrity: 100% Developed in Python for Python. - Navigation: Fast and easy to navigate. Helped by clear colored hyperlink styling. Especially the 'bread-crumb' -style path on top left of every page. Easy to see where you are and jump back up the tree. - Structure: The site is not a home but a super directory. I believe the 'Vaults of Parnassus' database only holds pointers to module home pages and a little metadata needed. Does not store sources. - Database: not known - Search: Search is quick and useful. Available 24/7 every page - URL accuracy: This one area the Vaults could be improved a little. Since the site is just a directory for many external websites, it is hard to catch and store a URL reference on the site itself. An obviously it is not intended to be used like that. For example, the GAMES page is http://py.vaults.ca/parnassus/apyllo.py/238168060 There's an entry "Crystal Space 0.14 - A free portable 3D engine written in C++, with python scripting support. <more>" When you roll the mouse over it shows the link for 'Crystal Space' as http://py.vaults.ca/apyllo2.py/66090028 but clicking on it will redirect your browser to its source forge home at http://crystal.sourceforge.net/ - Submit: Submit page is very simple. Nice option to autopost announcement to Python newsgroup. http://py.vaults.ca/submit/ - Community: One-to-many model. 'The Vaults' are the *essential* contributed software resource of the python universe. But apart from submission form, there is no direct community activity, threaded comment, notes of feedback etc. All that is done on comp.lang.python and/or emails to authors or in local forums organized by the authors on their own sites. - Design: Simple distinctive 'gothic' concept. Minimal graphics to keep it swift and focused, but a little design thought goes a long way. Very legible. - Printing: [important to test and consider this. sorry no printer available right now.] - Authoring: One person design built and maintains. - Favorite feature: 'Latest' tab at the top. Best way I know to find out anything that's hot AND usable in Python. - Performance: Usually fast and very well maintained. Periodically off-line. There were database/server access problems occasionally in the past. That has been fixed now I think. Occasionally referred off-site links are down or dead [not the fault of Parnassus]. Over time these are flagged and corrected. =============================================== ZOPE

Integrity: 100% Developed in Zope for Zope. - Navigation: fast with multiple links. Typically not more than 3 clicks away from anything. - Structure: Simple tabs/menu built around a growing cookbook metaphor with related news and fast multiple indexes based on category, age, and popularity. - Database: ZODB, Zope's own built-in object database. - Search: - URL accuracy: Good. By database id number. Zope publish everything in its ZODB as a URL. For example: http://www.zopelabs.com/cookbook/1014075524 - Community: Participatory. Strong community resource among several. Registration + membership login unlock submissions, comments etc. That applies an intrinsic set of Zope features for handling fine-grained control of logins, sessions and permissions - Design: slightly cluttered minimal graphics, but very rich fast access from multiple places to the same information. Nice colored syntax. [note: TO my knowledge, There are no beautiful looking Zope sites -- all built by ultrageeks for ubergeeks. Zope is not artist-friendly. If only.. ] - Printing: [important to test and consider this. sorry no printer available right now.] - Authoring: One person design built and maintains. Zope has step learning curve. The site is both a way to give back to the community and a hands-on experimental learning/development project. Active site-membership and strong mailing list helps this. Zope changes constantly, so does the engineering supporting the site. A good demonstration of Zope, which was designed to do just this kind of dynamic web publishing. - Performance: Good. =============================================== REBOL =============================================== http://www.reboltech.com/library/library.html Integrity: uncertain.. The website surely uses some REBOL but no indication of how and where is given. Even if all sort of cool REBOL scripts are used to generate these pages, the end result is very poor self presentation of REBOL capabilities. Missed opportunity! The good news - it's easily remedied. - Navigation: Very crude:-( Too many clicks. No feedback. Depends heavily on browser 'back' button. The library might benefit from a unique server alias, such as http://library.rebol.com - Structure: Flat HTML c.1993 Based on simple categories. - Database: not known - Search: None!!!! REBOL home page has a general search to get you one of many site pages, but nothing better focused than that for the library. Once on a page, you can use browser's own 'find' tool to zero in. - URL accuracy: Good. By regular text name. An old trusted advantage of raw minimal HTML site design is that every script can be referenced reliably by a unique static URL. Ironically, dynamic sites often have to work much harder to generate simple unique urls which can be reused by people for linking to their content. Here see for example: http://www.reboltech.com/library/html/timewebs.html or http://www.reboltech.com/library/scripts/timewebs.r That's great, because once found, you can tell someone fairly reliably where to get it another time. Unfortunately, there is no cross-link on the HTML version to the source file %timewebs.r - Submit: None. Since REBOL/View/Desktop is the proposed platform, the web presence for REBOL library is minimal. Through /View Desktop there is another community potential demonstrated, but still lacks essential features to be seriously useful. Both rebsites and the website lack feedback, search and versatile indexing tools. - Community: None. Lacks any on-site commentary, support, notes, stats or examples. Depends on mailing list. - Design: Clean minimal sans serif. Legible. More or less consistent with rebol docs and VID stuff. Black brown typography with duck-egg blue background highlight bars. - Printing: [important to test and consider this. sorry no printer available right now.] - Favorite feature: REBOL scripts are so small, and you can just cut'n'paste them right into the shell from a web page to run them. Yes! - Performance: Fast because it is simple. Your turn... - Integrity: - Navigation: - Structure: - Database: - Search: - URL accuracy: - Submit: - Community: - Design: - Printing: - Favorite feature: - Performance: ./Jason

Ditto from me. Andrew Martin -><- Andrew Martin

That's a good question. I hope the issue is still alive.

Oh the modesty! Ingo, your scripting ability is more advanced than you admit. REBOL's complexity seems to scale with your ability, so maybe you never quite feel like a guru. /// FAQ Writers, step forward by this Tuesday Oct 1 Thinking beyond the library, I propose to lead an effort to create a much richer FAQ for REBOL. I admire the Javascript FAQ at www.IRT.org ( http://developer.irt.org/script/script.htm ) and I think REBOL could benefit from a similar resource. In this vein, I've amassed roughly 1000 questions from messages on this list and other language FAQs. Most of the questions are newbie-level and have a web/HTML/Javascript perspective. They require clear, well thought-out answers. I've been disgracefully remiss about getting a posse together to tackle this. But shame has no value if it doesn't beget action. Any REBOLs with a flair for writing/editing documentation who would like to contribute time and expertise, please contact me directly no later than Tuesday, October 1. There is also a trusty-REBOLer with a hankering to write FAQ-builder script (output similar to IRT FAQ I mentioned). Gotta move fast, time is the enemy. Serious offers only. I will send out a reminder this Monday and start parcelling out work immediately. There will be a schedule, deadlines and a launch date, so no lollygagging! Pretty simple stuff, but long overdue. Regards, Ed

to contribute time and expertise,... Can you put it on a Wiki somewhere? That way it will become better edited and rewritten almost automatically, as people use it and say to themselves, I know the answer to that! or, "hang on, that's wrong, I'll just change it". Andrew Martin ICQ: 26227169 http://valley.150m.com/

Maybe. The first order of business is to do a good job answering the questions. We'll keep the data in a REBOL readable format so that we have lots of options.

Um... ideally that's what happens ;^). Hopefully we'll get a chance to see. For now, a basic, well written, well-organized FAQ is the goal. Thanks

Sorry for the late answer but I'm now in the process of covering my house Roof by myself since last thursday - so my time is accounted for by now and for the next 2 weeks before I return to some week-end study until mid-december. But since looking at this ML and learning REBOL are two of my favorite "rest areas" - I promise you that I will do all my possible to review and hopefully answer by myself as much as I can of your 1000 questions. I already have thought about some of them for myself - not couched on paper yet however - since I go quickly to the Web or to the REBOL console to answer them as soon as they arrive. As I also planned to start a new Tutorial/book about REBOL for starters who want to look at some first programming language using REBOL as their first tool - I didn't forget you Tim about what you are doing with hi-schoolers in Alaska and I'll also contact you as soon as possible - this will be of some usefulness to me and many others too. This the part I was missing the most when I was starting, even if there is much doc on the REBOL web site in many forms. But what I wanted to get at hand was some kind of "goal or task related" index and searching tool. The "how-to" were the most useful parts as much as were the official docs. The new official DOCs are missing too much examples to accompany the formal description of the REBOL WORDS. I even found that the format of the older Official DOCS - those on the accompanying REBOL for Dummies book - are better looking and easier to work with. May be it is a matter of style or presentation since when I looked back at the new official DOCs I found almost the same information. May be some examples missing but not much difference. I have yet to look back at the diffs. however. I hope this new FAQ arsenal will overcome this small hole in part as much as the planned REBOL SCRIPTS LIBRARY will help too - for starters to use some already "partly" tested REBOL code as a basis for their work. So if you want me to begin review your FAQ and submit some answers when it will be possible, simply send your list to me. I am working on a machine hosted by a Win98 OS. I'm waiting more input from you as the leader of this project. Also the idea of submitting any question to a wiki has some already tested virtue but any other way will do - like any Conference or Forum on /IOS or Yahoo Groups. I also plan to work on a derived "help add-on" that could quickly put Parens around the parameters found in any (or most simple) line of REBOL coding. This has been discussed here in some other thread (see refs. from Gregg and Ladislav) and is what also blocked me at first try from quickly reading some already existing REBOL scripts - well may be I would have not understood the algo. in the first reading but at least I would have skipped the search for the right number of arguments in the first place !!! - and this is some BAD REMARK that arise VERY OFTEN from Outsiders or newbies that give up after some try reading some REBOL code. This simple tool - that also could be used as part of some "pretty lister" - could save REBOLers this too frequent (but nevertheless REAL) comment about REBOL coding "laziness" (read one of the last refs. from Jason about the outsider's view of REBOL - from Python coders and the like!). Regards, Gerard P.S. I will do my best regarding the following but don't wait for me if I can't get at it quickly enough for the others to do the Job. I'll comment them for the second revision if any is needed ...

I agree entirely with you !!!