Talk:LocalTalk

Contradiction

I've noticed an item in two articles that contradict each other: AppleTalk says ("PhoneNet was considerably less expensive to install and maintain, and it is perhaps surprising that Apple did not move to this solution as well.") vs LocalTalk ("eventually Apple itself abandoned LocalTalk wiring and marketed PhoneNet as LocalTalk"). Can someone confirm if Apple ever offered a PhoneNet solution? --Steven Fisher 14:20, 17 May 2005 (UTC)

I think it's probably more accurate to say that they let traditional STP LocalTalk die out -- I don't believe they ever actually sold PhoneNet transceivers (Farallon created them, and they did eventually become something of a generic commodity). But by that time Ethernet was well on its way to its current market dominance -- I think LocalTalk as a network layer was all but dead by around 1994-5 when Apple started including AAUI and then 10Base-T ports on their hardware. The crossover cable killed ad hoc LocalTalk networking, and the elimination of the RS-422 serial port was just a shovel of dirt on the grave. The only place you still see it is in places where old 68K Macs (especially compact Macs) are recycled as terminals on small networks, and there aren't even all that many of those left. Haikupoet 18:21, 18 May 2005 (UTC)

I've updated the page. How does this look? --Steven Fisher 13:20, 23 May 2005 (UTC)

More LocalTalk vs. PhoneNet

The article curently contains the following language:

In addition to being lower cost, PhoneNet-wired networks were more reliable due to the connections being more difficult to accidentally disconnect. The improved reliability and lower cost led to PhoneNet largely supplanting LocalTalk wiring in low cost networking.

This struck me as funny because I always found Apple's locking LocalTalk DIN connectors to provide more reliable connections than the PhoneNet connectors. (Admittedly, my net was a mix of genuine Farralon PhoneNet and Belkin clones.) For me, the problem was "connectoritis"; the signal levels on the PhoneNet simply weren't large enough to overcome all the various oxides and the like that built up on the various connections in the PhoneNet system. (The Apple auto-terminating adapters were also far more clever than the PhoneNet plug-in termination resistors.)

Any other opinions? Should we moderate what sounds like POV to me?

Atlant 23:20, 16 July 2005 (UTC)

I maintained a large (~50 machine) multi-segment, multi-platform (Macs, PCs and DEC minis), physically large (entire floor of a hi-rise) PhoneNet network with piggybacked signals that were split out back-office to phones and network patchboards and never had any problems like this. On the other hand, the single LocalTalk box, attached to a PC and running directly to a router, was cause for all sorts of problems. YMMV. Maury Markowitz (talk) 17:19, 4 January 2009 (UTC)

Twisted Pair?

In several places in the article, it mentions using "twisted pair" for PhoneNet. I recall only ever using regular "silver satin" flat modular phone cord. The pairs were not twisted together like Cat3. (I don't recall, though, if the cables were reversing or straight-through.) Perhaps I was just carelessly using the wrong material (I did not know the difference back then).

Also, the following lines

It used standard unshielded twisted pair telephone wire with 6 position modular connectors (same as used in the popular RJ11 telephone connectors) connected to a PhoneNet transceiver

and

it used the "outer" pair of the modular connector

might cause confusion because they imply that PhoneNet used 6-conductor wiring. RJ11, RJ14, and RJ25 are the three configurations of the common 6-position modular plug and carry one, two, and three pairs, respectively. Single-line telephones commonly use RJ11; two-line telephones and, typically, key phones use RJ14 connectors. PhoneNet was a four-wire system and carried the LocalTalk signal on pair 2 of a RJ14, not pair 3 of a RJ25 (and certainly not on a RJ11). Starling2001 (talk) 22:05, 15 December 2009 (UTC)

There were a few inaccuracies in what I wrote above. Firstly, Cat3 is untwisted. Still, though, I believe PhoneNet used modular cord, not Cat3. Telephone wall-to-phone cords are usually reverse wired. But, as was the case three years ago, I still do not recall whether PhoneNet cables were reversed (the clip on both mod plugs on the same flat of the cable) or straight-through (one clip up, the other down).

Also, I indicated that key phones use RJ14 (the standard for 2-line phones). This is not quite accurate. Although often used (erroneously) to refer just to the physical form of a plug or jack, the RJ designation (such as RJ14 or RJ45) actually also specifies the particular arrangement of the electrical signals carried over each conductor. While the wiring typically used for key phones does, indeed, use the same 6P4C modular plug as used for the RJ14 found on 2-line phones, it is actually a different electrical standard and is designated RJ13.

Similarly, the 6P4C mod plug on PhoneNet is also the same plug as used for RJ13 and RJ14, but the wiring is electrically dissimilar. Thus, a PhoneNet jack is also not technically RJ14, even though the jack is physically identical. Even so, calling a 6P4C mod plug (as used for PhoneNet) an RJ14 is common enough usage that most manufacturers would even tend to recognize it as such.

The issue remains, though, that the article still refers to it as a RJ11 plug, which is a single-pair 6P2C plug and is definitely not compatible with PhoneNet.Starling2001 (talk) 01:56, 11 December 2012 (UTC)