AnsweredAssumed Answered

vrf multiple copies running

Question asked by VRFuser on Dec 16, 2001
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"><HTML><HEAD><META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1"><META NAME="Generator" CONTENT="MS Exchange Server version 5.5.2654.45"><TITLE>RE: vrf Multi-VEE-ing capability</TITLE></HEAD><BODY><P><FONT SIZE=2>Rob,</FONT></P><P><FONT SIZE=2>Actually, we found it does release it, but cannot reopen it. Regardless, we can't use it with another app that tries to open & close it. For that reason, we have started an evaluation of an alternate solution (i.e. not use VEE).</FONT></P><P><FONT SIZE=2>Paul H. Rubin</FONT><BR><FONT SIZE=2>Staff Engineer</FONT><BR><FONT SIZE=2>Test Engineering</FONT><BR><FONT SIZE=2>Ericsson Amplifier Technologies, Inc.</FONT><BR><FONT SIZE=2>49 Wireless Blvd.</FONT><BR><FONT SIZE=2>Hauppaugh, New York 11788</FONT><BR><FONT SIZE=2>Phone: (631) 357-8513</FONT><BR><FONT SIZE=2>E-Mail:</FONT></P><BR><BR><P><FONT SIZE=2>-----Original Message-----</FONT><BR><FONT SIZE=2>From: [<A HREF=""></A>]</FONT><BR><FONT SIZE=2>Sent: Monday, December 17, 2001 9:54 AM</FONT><BR><FONT SIZE=2>To: Paul Rubin (AMT)</FONT><BR><FONT SIZE=2>Subject: RE: vrf Multi-VEE-ing capability</FONT></P><BR><BR><P><FONT SIZE=2>I think it's a known VEE bug. VEE doesn't release the resource even</FONT><BR><FONT SIZE=2>if you execute a CLOSE. We don't even try to do this anymore.</FONT><BR><FONT SIZE=2>If pressed, however, we could funnel it through VEE via a TCP/IP socket</FONT><BR><FONT SIZE=2>or use two ports and a remote control switch |at least until this gets</FONT><BR><FONT SIZE=2>fixed>.</FONT></P><P><FONT SIZE=2>Rob</FONT></P><P><FONT SIZE=2>Rob Marquardt</FONT><BR><FONT SIZE=2>System Test Engineering</FONT><BR><FONT SIZE=2>Rockwell Automation / Allen-Bradley</FONT><BR><FONT SIZE=2>1201 So. 2nd Street</FONT><BR><FONT SIZE=2>Milwaukee, Wisconsin, 53204-2496, USA</FONT><BR><FONT SIZE=2>414-382-3098 Phone</FONT><BR><FONT SIZE=2>414-382-4820 FAX</FONT><BR><FONT SIZE=2></FONT></P><BR><BR><BR><BR><BR><P><FONT SIZE=2>"Paul Rubin (AMT)" |> on</FONT><BR><FONT SIZE=2>12/14/2001 03:53:10 PM</FONT></P><P><FONT SIZE=2>Sent by:</FONT></P><BR><P><FONT SIZE=2>To:   "''" |>,</FONT><BR><FONT SIZE=2></FONT><BR><FONT SIZE=2>cc:</FONT></P><P><FONT SIZE=2>Subject:  RE: vrf Multi-VEE-ing capability</FONT></P><BR><BR><P><FONT SIZE=2>Thomas,</FONT></P><P><FONT SIZE=2>Apparently you are only claiming that 2 _VEE_ programs can  "share" a Com</FONT><BR><FONT SIZE=2>port. Correct? Have you ever succeeded in having VEE  share a Com port with</FONT><BR><FONT SIZE=2>any other type application?</FONT></P><P><FONT SIZE=2>Paul H. Rubin</FONT><BR><FONT SIZE=2>Staff Engineer</FONT><BR><FONT SIZE=2>Test Engineering</FONT><BR><FONT SIZE=2>Ericsson Amplifier Technologies, Inc.</FONT><BR><FONT SIZE=2>49 Wireless Blvd.</FONT><BR><FONT SIZE=2>Hauppauge, New York 11788</FONT><BR><FONT SIZE=2>Phone: (631)  357-8513</FONT><BR><FONT SIZE=2>E-Mail:</FONT></P><P><FONT SIZE=2>-----Original Message-----</FONT><BR><FONT SIZE=2>From: Thomas Vauderwange  [<A HREF=""></A>]</FONT><BR><FONT SIZE=2>Sent: Friday, December 14, 2001  4:35 PM</FONT><BR><FONT SIZE=2>To:</FONT><BR><FONT SIZE=2>Subject: Re: vrf  Multi-VEE-ing capability</FONT></P><BR><P><FONT SIZE=2>Hi Paul,</FONT></P><P><FONT SIZE=2>Well, found it hard to believe when I heard it. I was aware that VEE  won't</FONT><BR><FONT SIZE=2>let the COM-port loose, once it got hold of it, so the idea to try</FONT><BR><FONT SIZE=2>COM-port-sharing with non-VEE-apps was abandoned.</FONT></P><P><FONT SIZE=2>Then, that customer from over there in luvly North Carolina called one</FONT><BR><FONT SIZE=2>day, reporting a 'severe firmware bug' in our machine's microcontroller, as</FONT><BR><FONT SIZE=2>there were the two LEDs that indicate the two possible control modes of</FONT><BR><FONT SIZE=2>that  machine blinking! In a very regular manner, the right one - the left</FONT><BR><FONT SIZE=2>one. As  something like this was completely unheard of, I started the</FONT><BR><FONT SIZE=2>investigation  and found out that my software, which he uses to</FONT><BR><FONT SIZE=2>control/monitor the machine  from a PC had been started two times</FONT><BR><FONT SIZE=2>simultaneously on that PC. On the copy  in the foreground, that customer</FONT><BR><FONT SIZE=2>had pushed the button that did send (via  COM1) the signal to change</FONT><BR><FONT SIZE=2>control mode. The other instance did notice the  deviation with its next</FONT><BR><FONT SIZE=2>data exchange and thus ordered the machine back in  the other mode. Which</FONT><BR><FONT SIZE=2>led the copy in front of the screen to send another  command...</FONT></P><P><FONT SIZE=2>Anyway, it worked fine. Unfortunately, you need a real one of these</FONT><BR><FONT SIZE=2>machines to try my software out, which would otherwise be available as a</FONT><BR><FONT SIZE=2>free download from the company's website. But, in case you happen to need a</FONT><BR><FONT SIZE=2>heater-chiller circulator, then that company would be the right address.</FONT></P><P><FONT SIZE=2>Thomas</FONT></P><P><FONT SIZE=2>"Paul Rubin (AMT)" schrieb:</FONT></P><P><FONT SIZE=2>Thomas,</FONT></P><P><FONT SIZE=2>I'm quite interested in your comment about RS232 ports.  We have not been</FONT><BR><FONT SIZE=2>able to resolve a problem with 2 applications trying to  use the same Com</FONT><BR><FONT SIZE=2>port. While we can "Close" the port from the  Vee application, there</FONT><BR><FONT SIZE=2>appears to be no way to reopen it after the other  application (NOT VEE)</FONT><BR><FONT SIZE=2>releases it. I imagine it would be possible to  replicat the other</FONT><BR><FONT SIZE=2>application in VEE, if absolutely necessary, but that  would be a</FONT><BR><FONT SIZE=2>significant undertaking as there are a number of different  combinations</FONT><BR><FONT SIZE=2>possible.</FONT></P><P><FONT SIZE=2>Please advise</FONT></P><P><FONT SIZE=2>Paul H. Rubin</FONT><BR><FONT SIZE=2>Staff  Engineer</FONT><BR><FONT SIZE=2>Test Engineering</FONT><BR><FONT SIZE=2>Ericsson Amplifier Technologies, Inc.</FONT><BR><FONT SIZE=2>49 Wireless Blvd.</FONT><BR><FONT SIZE=2>Hauppaugh, New York  11788</FONT><BR><FONT SIZE=2>Phone: (631) 357-8513</FONT><BR><FONT SIZE=2>E-Mail:</FONT></P><BR><P><FONT SIZE=2>-----Original Message-----</FONT><BR><FONT SIZE=2>From: Thomas Vauderwange [<A HREF=""></A>]</FONT><BR><FONT SIZE=2>Sent: Friday, December 14, 2001 4:04 PM</FONT><BR><FONT SIZE=2>To:</FONT><BR><FONT SIZE=2>Subject: vrf Multi-VEE-ing capability</FONT></P><P><FONT SIZE=2>Hi Shawn,</FONT></P><P><FONT SIZE=2>Indeed it is no problem at all to have several  different VEE programs</FONT><BR><FONT SIZE=2>running on</FONT><BR><FONT SIZE=2>the same PC  the same time. Even tricky stuff, like RS232-comms etc. My</FONT><BR><FONT SIZE=2>customers</FONT><BR><FONT SIZE=2>and the folks down in the production  hall do this all the time with my</FONT><BR><FONT SIZE=2>software</FONT><BR><FONT SIZE=2>(Win98, NT4, W2K), and everybody's happy. This happened when I  changed</FONT><BR><FONT SIZE=2>from good</FONT><BR><FONT SIZE=2>ol' VEE3-interpreter (where  it did not work) to the one that came with</FONT><BR><FONT SIZE=2>VEE5.</FONT><BR><FONT SIZE=2>And: someone found out by coincidence by just mistakenly  starting several</FONT><BR><FONT SIZE=2>processes. Even sharing one  and the same COM-port (!!) worked without</FONT><BR><FONT SIZE=2>having</FONT><BR><FONT SIZE=2>been specified!</FONT><BR><FONT SIZE=2>All I had to do to make  this a real feature is to specify different</FONT><BR><FONT SIZE=2>working</FONT><BR><FONT SIZE=2>directories where different config-files could be placed. And,  of course,</FONT><BR><FONT SIZE=2>to</FONT><BR><FONT SIZE=2>allow for an identifier string in  the left top of the temperature graph,</FONT><BR><FONT SIZE=2>because</FONT><BR><FONT SIZE=2>otehrwise nobody could tell the processes apart...</FONT></P><P><FONT SIZE=2>Regards,</FONT></P><P><FONT SIZE=2>Thomas</FONT></P><P><FONT SIZE=2>Shawn Fessenden schrieb:</FONT></P><P><FONT SIZE=2>> Well thank you everybody for your comments!</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> There are some real  eye-openers there, especially Jens. While I strongly</FONT><BR><FONT SIZE=2>> disagree with VEE's affinity for "structured"  code, I suspect that's</FONT><BR><FONT SIZE=2>simply</FONT><BR><FONT SIZE=2>> because we  have different ideas of what structured code is but, "that's</FONT><BR><FONT SIZE=2>> probably a discussion more suited to a  comp.programming group," (he</FONT><BR><FONT SIZE=2>said,</FONT><BR><FONT SIZE=2>>  trying to stay on topic).</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> On the other side of the fence, VEE's innate flowchart-like  diagramatic</FONT><BR><FONT SIZE=2>/</FONT><BR><FONT SIZE=2>> execution model is  definitely a godsend. In fact, whenever customers</FONT><BR><FONT SIZE=2>> explicitly request VEE this model is almost always the  reason why. When</FONT><BR><FONT SIZE=2>I</FONT><BR><FONT SIZE=2>> first started using  VEE, I just assumed customers wanted VEE because it</FONT><BR><FONT SIZE=2>was</FONT><BR><FONT SIZE=2>> an HP product.</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> Anyway Jens, I found this particularly  interesting:</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>>  >2 VEE clients and the VEE-controller</FONT><BR><FONT SIZE=2>>  >can run on the same PC or on different PCs.</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> If I understand this  correctly you can run *three* instances of VEE on</FONT><BR><FONT SIZE=2>*one*</FONT><BR><FONT SIZE=2>> computer? If this is the case, then my hat is off to you  and I will</FONT><BR><FONT SIZE=2>> henceforth keep my mouth  shut!</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> Using  NT 4.0 I've had terrible problems just keeping *one* instance</FONT><BR><FONT SIZE=2>going  on</FONT><BR><FONT SIZE=2>> anything less than a PII 233! At the  risk of boring everybody, let me</FONT><BR><FONT SIZE=2>> outline  two of these problems. I really need to understand why it is I</FONT><BR><FONT SIZE=2>have</FONT><BR><FONT SIZE=2>> such awful problems and others do  not. At the time, we had HP field</FONT><BR><FONT SIZE=2>>  application engineers crawling all over what I did and their conclusion</FONT><BR><FONT SIZE=2>was</FONT><BR><FONT SIZE=2>> "It's fine, that's just the way  it works". Ever since then I've been</FONT><BR><FONT SIZE=2>saying,</FONT><BR><FONT SIZE=2>> "well then it doesn't work right".</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> These two problems  are actually just different aspects of the same root</FONT><BR><FONT SIZE=2>> cause, CPU hogging:</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> 1. Printing. Customer wants a simplistic graph of  readings over time.</FONT><BR><FONT SIZE=2>Using</FONT><BR><FONT SIZE=2>> the Print Screen  object, it took over 45 minutes to print one graph.</FONT><BR><FONT SIZE=2>This is</FONT><BR><FONT SIZE=2>> over a 100MB network. Spooler priority was set  highest. I abandoned the</FONT><BR><FONT SIZE=2>> spooler and of  course that speeded up the printing to a minute or two,</FONT><BR><FONT SIZE=2>but</FONT><BR><FONT SIZE=2>> the UI completely stops responding (of course)  and that's unacceptable.</FONT><BR><FONT SIZE=2>I</FONT><BR><FONT SIZE=2>> wound up using  Microsoft's DIBAPI to capture the screen myself & print</FONT><BR><FONT SIZE=2>it.</FONT><BR><FONT SIZE=2>> The catch was that this utility had to shift  itself to</FONT><BR><FONT SIZE=2>HIGH_PRIORITY_CLASS</FONT><BR><FONT SIZE=2>> to produce a  graph in under 5 minutes. This slowed down the UI some, but</FONT><BR><FONT SIZE=2>the</FONT><BR><FONT SIZE=2>> customer accepted the result as a necessary  trade-off.</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>>  2. Browsing. The test scenario is a room full of some 100 UUTs. Test</FONT><BR><FONT SIZE=2>time</FONT><BR><FONT SIZE=2>> varies from 4 hours to three or four  days sometimes. Data points consist</FONT><BR><FONT SIZE=2>of</FONT><BR><FONT SIZE=2>>  samples of 6 channels every 30 seconds. The DAC box is an Agilent 34970,</FONT><BR><FONT SIZE=2>one</FONT><BR><FONT SIZE=2>> per UUT. Loading the test line is  pretty simple. One plug for power, one</FONT><BR><FONT SIZE=2>for</FONT><BR><FONT SIZE=2>>  4 data channels. The other two channels are from sensors placed inside</FONT><BR><FONT SIZE=2>the</FONT><BR><FONT SIZE=2>> UUT by test personnel. A bar code is  scanned and the test cycle begins.</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> Now, the people who run the test system are union  laborers, not</FONT><BR><FONT SIZE=2>engineers.</FONT><BR><FONT SIZE=2>> There are usually  three people per test room. It takes all of about an</FONT><BR><FONT SIZE=2>hour</FONT><BR><FONT SIZE=2>> to get a full room under way and then they sit  back and wait. It's no</FONT><BR><FONT SIZE=2>fun</FONT><BR><FONT SIZE=2>> sitting there  staring at a graph or a summary of all running tests, so</FONT><BR><FONT SIZE=2>they</FONT><BR><FONT SIZE=2>> usually start playing solitare or browsing the  web, looking for</FONT><BR><FONT SIZE=2>girly-pix.</FONT><BR><FONT SIZE=2>> Now granted  that this isn't the ideal scenario, but realistically it's</FONT><BR><FONT SIZE=2>to be</FONT><BR><FONT SIZE=2>> expected. As soon as VEE is put in the  background, problems develop.</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> In the first place, after about 40 UUTs VEE  starts having a hard time</FONT><BR><FONT SIZE=2>> keeping up with  sampling data, updating 34970 displays, checking for</FONT><BR><FONT SIZE=2>card</FONT><BR><FONT SIZE=2>> swaps or entire 34970 swaps, verifying 34970  configuration, analyzing</FONT><BR><FONT SIZE=2>data</FONT><BR><FONT SIZE=2>> to decide on  shifts in test phases, testing com ports for proper</FONT><BR><FONT SIZE=2>> communication, watchdog pinging other test computers to  make sure</FONT><BR><FONT SIZE=2>they're</FONT><BR><FONT SIZE=2>> ok, etc. This is even  with VEE in the bloody FOREGROUND. When it gets</FONT><BR><FONT SIZE=2>put in</FONT><BR><FONT SIZE=2>> the background, look out!</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> Now I realize full well that  I'm no VEE expert. Like I've said, 30 years</FONT><BR><FONT SIZE=2>of</FONT><BR><FONT SIZE=2>> programming doesn't mean a thing when it comes to VEE so we  called in HP</FONT><BR><FONT SIZE=2>> experts for expert advice.  After over half a year of tweaking and some</FONT><BR><FONT SIZE=2>> truely extraordinary efforts on the part of various HP  engineers, the</FONT><BR><FONT SIZE=2>> customer was still  dissatisfied and started threatening to pull out, and</FONT><BR><FONT SIZE=2>for</FONT><BR><FONT SIZE=2>> that I don't blame them. They dumped *huge*  amounts of cash on hardware</FONT><BR><FONT SIZE=2>and</FONT><BR><FONT SIZE=2>> they  couldn't even run at 50% capacity.</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> Even then, memory leaks caused by the infamous  error pins would, after</FONT><BR><FONT SIZE=2>only</FONT><BR><FONT SIZE=2>> a few days of  continuous operation, cause VEE to reset it's heap (or</FONT><BR><FONT SIZE=2>> something) and all in-progress tests were lost. Granted,  there was</FONT><BR><FONT SIZE=2>always a</FONT><BR><FONT SIZE=2>> current status backup  stored on the drive, but this was updated every</FONT><BR><FONT SIZE=2>time</FONT><BR><FONT SIZE=2>> data changed on any DAC box and the moment VEE lost it's  variables this</FONT><BR><FONT SIZE=2>file</FONT><BR><FONT SIZE=2>> was overwritten with  emptyness. The individual data files themselves</FONT><BR><FONT SIZE=2>> remained, but without status information (current test  phase, phase test</FONT><BR><FONT SIZE=2>> instructions, current  machine state, state machine registers, etc.) the</FONT><BR><FONT SIZE=2>> state machine couldn't possibly continue.</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> It was finally decided to  move to a compiled language so I locked myself</FONT><BR><FONT SIZE=2>in</FONT><BR><FONT SIZE=2>> a closet for 14 weeks and re-wrote the entire thing in C++.  And when I</FONT><BR><FONT SIZE=2>say</FONT><BR><FONT SIZE=2>> re-wrote, I mean exactly  that. The C++ version is almost an exact</FONT><BR><FONT SIZE=2>duplicate</FONT><BR><FONT SIZE=2>> of the VEE version. VEE's failures in this instance were  definitely not</FONT><BR><FONT SIZE=2>due</FONT><BR><FONT SIZE=2>> to software  architecture. This application is currently buzzing happily</FONT><BR><FONT SIZE=2>> away, testing like crazy. There are no issues  with printing graphs,</FONT><BR><FONT SIZE=2>playing</FONT><BR><FONT SIZE=2>> solitare or  browsing the web.</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> The *opinions* I express here are based largely on this one</FONT><BR><FONT SIZE=2>extraordinarily</FONT><BR><FONT SIZE=2>> negative experience. This  was definitely the most detailed analysis I've</FONT><BR><FONT SIZE=2>> ever made of VEE on Windows in the wild, though I've had  plenty of</FONT><BR><FONT SIZE=2>somewhat</FONT><BR><FONT SIZE=2>> negative experiences  with it before. The icing on the cake was HP's own</FONT><BR><FONT SIZE=2>> conclusion that "yes, everything is fine. That's the  way it works".</FONT><BR><FONT SIZE=2>Since</FONT><BR><FONT SIZE=2>> then I've  recommended to those customers who want to use VEE that they</FONT><BR><FONT SIZE=2>then</FONT><BR><FONT SIZE=2>> should stick with HP-UX. Sometimes that's fine,  and sometimes it's not</FONT><BR><FONT SIZE=2>> because Windows is  a requirement. In that case, we use "Windows</FONT><BR><FONT SIZE=2>specific"</FONT><BR><FONT SIZE=2>> tools like Visual Basic, Visual C++, Delphi or  C++ Builder.</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>>  Now I challenge any of you to put yourself in my shoes: Put your life</FONT><BR><FONT SIZE=2>on</FONT><BR><FONT SIZE=2>> hold for 14 weeks of consecutive 16-20  hour days WITHOUT PAY, bear the</FONT><BR><FONT SIZE=2>whip</FONT><BR><FONT SIZE=2>> from  management AND the customer, watch HP go over everything you've</FONT><BR><FONT SIZE=2>done</FONT><BR><FONT SIZE=2>> with a fine-tooth comb and shrug  their shoulders, watch five years of</FONT><BR><FONT SIZE=2>your</FONT><BR><FONT SIZE=2>>  life helping to build a company from a two-man operation to a small</FONT><BR><FONT SIZE=2>global</FONT><BR><FONT SIZE=2>> enterprise go down the drain as law  suits and bankruptcy ensue and then</FONT><BR><FONT SIZE=2>have</FONT><BR><FONT SIZE=2>>  anything positive at all to say about VEE. Here it is, on public</FONT><BR><FONT SIZE=2>record:</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> IN  SPITE OF ALL OF THIS I STILL LIKE VEE!!!</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>> So please, if anybody has  anything to say to me personally about VEE</FONT><BR><FONT SIZE=2>bashing</FONT><BR><FONT SIZE=2>> then lets keep it off the list as it probably won't be  productive at all</FONT><BR><FONT SIZE=2>and</FONT><BR><FONT SIZE=2>> I guarantee I'll  respond to all flames with a blow-torch.</FONT><BR><FONT SIZE=2>>  -SHAWN-</FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>>  _________________________________________________________________</FONT><BR><FONT SIZE=2>> Join the world's largest e-mail service with MSN  Hotmail.</FONT><BR><FONT SIZE=2>> <A HREF="" TARGET="_blank"></A></FONT><BR><FONT SIZE=2>></FONT><BR><FONT SIZE=2>>  ---------------------------------------------------------------------</FONT><BR><FONT SIZE=2>> This is the "vrf" maillist, managed by  Majordomo. To send messages to</FONT><BR><FONT SIZE=2>> this  maillist, just email to "".  Subscriptions and</FONT><BR><FONT SIZE=2>> unsubscriptions are done  through the address</FONT><BR><FONT SIZE=2>"".</FONT><BR><FONT SIZE=2>> If you need details, just send a message  containing the text "help"</FONT><BR><FONT SIZE=2>> to  "".</FONT><BR><FONT SIZE=2>>  ---------------------------------------------------------------------</FONT></P><P><FONT SIZE=2>---------------------------------------------------------------------</FONT><BR><FONT SIZE=2>This is the "vrf" maillist, managed by  Majordomo. To send messages to</FONT><BR><FONT SIZE=2>this  maillist, just email to "".  Subscriptions and</FONT><BR><FONT SIZE=2>unsubscriptions are done  through the address</FONT><BR><FONT SIZE=2>"".</FONT><BR><FONT SIZE=2>If you need details, just send a message containing  the text "help"</FONT><BR><FONT SIZE=2>to  "".</FONT><BR><FONT SIZE=2>---------------------------------------------------------------------</FONT></P><BR><BR></BODY></HTML>