Talk:Asterisk (PBX)

From Wikipedia, the free encyclopedia

Contents 1 On Portal:Free software, Asterisk PBX is current featured article 2 A lot like assembler? 3 Recommend Removal Of Extraneous External Links 4 Additional Link Discussion 5 Recommend Removal Of Criticism section 6 MozIAX

[edit] On Portal:Free software, Asterisk PBX is current featured article

Just to let you know. The purpose of featuring an article is both to point readers to the article and to highlight it to potential contributors. It will remain the feature for a week or so. Gronky 14:16, 27 January 2006 (UTC)

asterisk is dual licensed. gpl license, and digium commercial license (similar to mysql's dual license). only mentioning the gpl license is insufficient and incomplete. —The preceding unsigned comment was added by 66.118.69.158 (talk • contribs) 02:51, February 1, 2006.

[edit] A lot like assembler?

I am removing the following text discussing the dialplan language:

"The language is very primitive -- somewhat like assembler."

Whether a language is "primitive" or not seems like a pretty ambiguous question. However, having used various assembly languages and having programmed several Asterisk dialplans, I can say for certain that the Asterisk dialplan language is nothing like assembler. DrDeke 21:59, 24 March 2006 (UTC)

Having used many types of assembler/assembly (PowerPC, SHARC, i386, x86_64, MIPS, Alpha) and observed many others (680x0, IA-64, SPARC, S/390...), I can say for certain that Asterisk dialplan language is disturbingly similar to assembler. I'd describe the Asterisk dialplan language as a bastard child of Perl and the Windows *.ini file, but "like assembler" is not a terrible description. There is something terribly wrong with the Asterisk dialplan language. The syntax is very irregular. The one language behaves in multiple ways, without any clear distinction. Straying so far from the ALGOL-style syntax (as used by C/C++/Java/C#, Ada/VHDL, Pascal/Modula2, etc.) is just cruel. The world doesn't need another abomination like make or m4. (I'll excuse FORTRAN, as it predates ALGOL) The Backus-Naur form and lexer for Asterisk dialplan language would be interesting at least, and possibly terrifying. AlbertCahalan 04:16, 22 April 2006 (UTC)

Would I be correct in assuming that neither of you have made use of Asterisk Extention Language (AEL)? —The preceding unsigned comment was added by 68.209.203.224 (talk • contribs) 12:04, April 23, 2006.

Probably so, because AEL is a new feature. It doesn't look all that different than the old stuff though. Check out this article where they say that "The opening curly-brace must appear as above. Moving it to the following line may have disastrous consequences!". (they also say "Please note that AEL is still considered EXPERIMENTAL at this time.") What a joke. Why have the C-style curly braces if whitespace is still critical? I like the way C works, and I think I'd like the way Python works (no curly braces, and whitespace is meaningful), but this is neither. Since the braces are already there, switching to a C-style syntax is probably the better route. A lone semicolon could be made to do nothing, allowing for the semicolons to be optional after the closing curly braces. AlbertCahalan 20:03, 23 April 2006 (UTC)

It seems that AEL2 fixes many of the AEL problems. That's even more experimental, and not even available in any released version of Asterisk. Progress is being made. There is a BNF now! Still, AEL2 is not a current Asterisk feature. It might be in a future release. Why is the language part case-sensitive and part not? Another good one: "Every character enclosed by the switch expression's parenthesis are included verbatim in the labels generated. So watch out for spaces!" More inconsistancy: "AEL2 introduces the special label "1", which is the beginning context number for most extensions." Well, I'm very glad to see things getting better. AEL2 is no worse than Perl. (in all seriousness) We can all celebrate when it finally gets released. Hopefully someday an AEL3 will be done, as clean as Java. AlbertCahalan 20:03, 23 April 2006 (UTC)

As you've got terribly strong feelings about this, then you should probably have a conversation with Steve Murphy, who's doing the AEL2 work. Developers' conference calls take place on Tuesdays. —The preceding unsigned comment was added by 68.209.203.224 (talk • contribs) 13:32, April 23, 2006.

[edit] Recommend Removal Of Extraneous External Links

This article is starting to look like a Link Farm with all the external links at the bottom. Wikipedia standards WP:EL state:

"An article about any organization, person, web site, or other entity should link to that entity's official site, if there is one."

The other link information can be found on the official website, is redundant and should be removed. I suggest removal of all links other than official website or open an WP:RFC for discussion. Calltech 15:20, 13 November 2006 (UTC)

I agree, I have cleaned up the links in the past but did not go as far to remove them all. BJ^Talk 18:44, 13 November 2006 (UTC)

External links removed and RfC opened, 14 November 2006. Calltech 14:30, 16 November 2006 (UTC)

Until there are people disputing this on the talk page, I don't think an RfC is necessary. It is hard for us to express a "third opinion" when there are no second opinions being expressed. I think you should close the RfC until you have discussed it here more. If no one wants to dispute your removals, then go right ahead. Grouse 17:41, 27 November 2006 (UTC)

Closed RfC. Calltech 18:40, 27 November 2006 (UTC)

[edit] Additional Link Discussion

Two additional links were added that need discussion. They both point to apparently the same new website that is under construction by Asterisk (AsteriskNow). This site requires registration and login. Also, the official website lists this site right upfront, which means the new links add nothing new and are therefore redundant. Recommend removal of both. Calltech 13:26, 8 December 2006 (UTC)

Oops, I was the one to take the 'See also' and move it to 'Official website' to get it in the right place but forgot to remove the one from 'See also' (fixed). I think a tidbit and the link should be placed in 'Distributions' and removed from 'Official website' BJ^Talk 14:17, 8 December 2006 (UTC)

I agree. Perhaps listing the site is a bit early since its in beta and under construction, but moving to Distributions makes sense. Thanks for your contributions. Calltech 17:09, 8 December 2006 (UTC)

[edit] Recommend Removal Of Criticism section

The criticism section really doesn't permit a balanced view to be presented. It contains a very one-sided criticism of the architecture and licensing of Asterisk, without the counter viewpoint. Given that it is, at best, opinion, not fact, it has no place in an encyclopedia entry. Corydon76 08:51, 14 January 2007 (UTC)

Well I was going to revert your edit but the criticisms section cited no sources. I will try to find some sources and rebuild a smaller version of the section. BJ^Talk 09:06, 14 January 2007 (UTC)

Actually, what would be preferable would be to recategorize some of the criticisms into other sections, as long as the criticisms were each matched by the counter viewpoint, in order to maintain a NPOV. For example, a section labelled Licensing might describe the dual license of Asterisk, while pointing out that others disagree with maintaining a dual license and either listing their reasons or linking to a page describing the arguments for and against dual licenses. Corydon76 09:27, 14 January 2007 (UTC)

"NPOV" doesn't mean "state a counter for every argument". In this case, I think we need to write what's written elsewhere (with citations); Asterisk has dual license, Asterisk needs a timing source that the Digium boards can supply, and Asterisk's architecture doesn't allow plug-in modules. If specific people have argued specific things about these facts, that can be mentioned. But people need to have the facts, not conclusions drawn by the Wikipedia editors. --Alvestrand 10:40, 14 January 2007 (UTC)

But "NPOV" requires that the mainstream viewpoint be given appropriate weight. The problem with this section was that a minority viewpoint was presented, without presenting the mainstream viewpoint. Finally, I'm not sure where you get the idea that Asterisk doesn't allow plug-in modules. There are over 130 plugin modules in the base install and there is no limit imposed on how many additional modules may be loaded. Corydon76 15:04, 14 January 2007 (UTC)

That's how I interpreted the removed "criticisms" section, but I'm happy that I was wrong. --Alvestrand 21:46, 14 January 2007 (UTC)

For one thing, Asterisk does not need the timing source from Digium cards. It can use them or other brands of cards, it can use the timing source provided by their dummy driver (free) or if you are using Linux, the 2.4 kernel has the timing built in. As far as the licence, How about stating the reason for the dual licence, instead of just outright denouncing it? Something like "the commercial version includes commercial codecs that cannot be included in the opensource version". Just stating what is written elsewhere doesn't make it NPOV; if I put on my own webpage that the Asterisk is the best thing ever and then cite it as a source, doesn't make it fact. --132.20.251.4 07:06, 17 January 2007 (UTC)

[edit] MozIAX

I added a link to MozIAX, a Mozilla Firefox extension for Asterisk. That article is just a little stub and might can merge and redirect here. Anyway there it is. By the way, We have a new VoIP group at Wikiversity and are playing with Skype, MozIAX and other VoIP stuff. • Q^#o • 02:32, 23 January 2007 (UTC)

I wonder if it might be better to link to the IAX2 entry and link MozIAX from there. Certainly MozIAX would run with anything implementing the IAX2 protocol and is therefore not Asterisk-specific. Corydon76 05:52, 30 January 2007 (UTC)