INFO-VAX Sat, 14 Jun 2008 Volume 2008 : Issue 330 Contents: Re: Chuck Norris says "Drill now in America for oil" Re: Chuck Norris says "Drill now in America for oil" Re: Chuck Norris says "Drill now in America for oil" Re: Chuck Norris says "Drill now in America for oil" Re: Chuck Norris says "Drill now in America for oil" Re: Chuck Norris says "Drill now in America for oil" Re: Chuck Norris says "Drill now in America for oil" Does anyone know if and how VMS figured in this? http://vmspython.dyndns.org/DownloadAndInstallationPython Re: http://vmspython.dyndns.org/DownloadAndInstallationPython Re: http://vmspython.dyndns.org/DownloadAndInstallationPython Re: http://vmspython.dyndns.org/DownloadAndInstallationPython Re: http://vmspython.dyndns.org/DownloadAndInstallationPython Re: Interesting job ad from HP Re: Interesting job ad from HP Re: Interesting job ad from HP Re: Interesting job ad from HP Re: Interesting job ad from HP Re: Interesting job ad from HP Re: Interesting job ad from HP Re: Interesting job ad from HP Re: Interesting job ad from HP Re: Interesting job ad from HP Re: Interesting job ad from HP OpenVMS Alpha 8.4 release date and new functionality ? Re: OpenVMS Alpha 8.4 release date and new functionality ? Re: OpenVMS Alpha 8.4 release date and new functionality ? Re: OpenVMS for Small and Medium Businesses? Re: PCL and PDF support in DCPS. Re: Python for VMS Re: TCPware 5.8 and MIME What happened to VAXeln after DEC stopped selling it ? Re: What happened to VAXeln after DEC stopped selling it ? Re: What happened to VAXeln after DEC stopped selling it ? What is a Weendoze .CHM file? Re: What is a Weendoze .CHM file? Re: What is a Weendoze .CHM file? Re: What is a Weendoze .CHM file? Re: What is a Weendoze .CHM file? Re: What is a Weendoze .CHM file? ---------------------------------------------------------------------- Date: Fri, 13 Jun 2008 14:16:03 -0400 From: JF Mezei Subject: Re: Chuck Norris says "Drill now in America for oil" Message-ID: <4852bb3a$0$31210$c3e8da3@news.astraweb.com> Dan O'Reilly wrote: > I just can't let this one pass. Neither can I. > When the Alaska Pipeline was built, the > environmentalists wrung their hands in sorrow for the herds of caribou that > were going to die off because we've disturbed their environment - yet > science has shown that the pipeline has directly contributed to their > INCREASE in population. In 1989, 90 and 91, I bicycled up the Dempster highway to Inuvik. (near coast of arctic ocean). This highway was started in the late 1950s and finished in 1979. This is a dirt road, but maintained to high grade. At one point on the highway, you rise to a high altitude on a plateau. From there you can see "roads" leading to many places. They are not roads. They are the leftovers from trucks that traveled on the toundra for oil exploration BACK IN THE 1950s. 40 years later, the damage done by one truck riding on the toundra is still clearly visible. That is how fragile the nature is there. What you see as a bush may in fact be a 100 year old tree. Now, the truly fagile environments start about 2/3 of the way between anchorage and prudhoe bay. (in canadian terms, it starts at about Dawson City). The porcupine cariboo don't travel as far west as the current alaska pipeline. But the new oil fields Bush so desperatly wants to develop are in their path, right next to the canadian border, IN AN INTERNATIONAL NATURE PRESERVE that was created to protect the very fragile wildlife ecosystem. Don't think for a minute that the development of oil fields won't seriously damage the area. You don't just deposit an oil rig on the ground from a helicopter. You need to build roads, bring in thousands of trucks, you need quaries to crush tons and tons of rocks. And the roads, while unpaved, are actually high tech because you need to insulate the road from the ground because if the ground melts below the road, the ground sinks and the area becomes a lake. > It's the same thing with nuclear power. People live in the 70's with that > too. The types of reactors in "The China Syndrome" aren't even built > anymore, they've been far surpassed by better, safer technology. Humm, when was the last truly new nuclear reactor built (and put into commercial operation) in the USA ? > Wind power and solar power are pipe dreams; they're unrealistic in what > they provide for the resources (for example, vast land acreages) they > consume and their cost. Exageration. Until we can get fusion (and matter/anti matter reactors from Star Trek) working, there is not going to be ONE solution to our energy needs. There will be many solutions, each contributing their little bit to the goal. Say you need to reduce your oil consumption by 30%. You could get 10% from solar, 10% from wind and 10% from conservation (increased efficiency, decreased consumption (etter home isulation, energy efficient lighthing etc). Solar can be on a large scale, or it can be in small scale. Having solar panels to heat your water for your home will greatly reduce your energy bill. > Biofuels are turning out to be one of the worst > mandates in many years by the governments of the world No. The current brou-ha-ha is of a political nature. Essentially the oil companies battling against ADM in the USA. Don't think for a minute that biofuel production capacity magically increased tenfold this year to suddently start to debalance world food prices. For instance, biofuel production in the USA doesn't have any relationship with rice production in thailand. Consider that the USA had a huge surplus capacity. This is one reason the USA is fighthing so hard to retain the highest subsidies in the world to its farmers. How can a country instantly move from having a production overcapacity to being at a stage where there would be a production deficit causing price increases ? This is all speculation driven. The only thing that has changed is that farmers now must pay much higher fuel prices for their tractors. And guess what, the day the USA (and France) stop subsidizing their farmers, Africa will suddently become far better off. By dumping their surplus production to Africa at low prices (or for free), it makes it impossible for farmers in africa to run a competitive business. A large number of nations have been demanding the USA and France stop subsidizing their farmers for a number of years. ------------------------------ Date: Fri, 13 Jun 2008 14:31:07 -0400 From: JF Mezei Subject: Re: Chuck Norris says "Drill now in America for oil" Message-ID: <4852bec3$0$20616$c3e8da3@news.astraweb.com> ultradwc@gmail.com wrote: > few days worth of oil? a spokesman from exxon just did > an interview on fox Fox is a republican entertainment network. Its onwer has publically admitted to meddling in editorial content and instructing its various media to not discuss global warming in a positive way. > and said 68 BILLION barrels of oil are > sitting in the gulf, alaska and the west ... that is hardly a > few days worth From: https://www.cia.gov/library/publications/the-world-factbook/print/us.html Oil - production: 8.322 million bbl/day (2005 est.) Oil - consumption: 20.8 million bbl/day (2005 est.) Ok, so I'll concede... more than 3 days. It is 3.26 days. > rigs fail now, they do not dump oil and damage anything ... > you are still living in the past ... time to get to speed with > the technology ... OK, lets discuss how ships fail. Are you aware that there is no pipeline between alaska and mainland USA and it is all shipped in super tankers. Do you remember Exxon Valdez by any chance ? ------------------------------ Date: 13 Jun 2008 13:49:05 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Chuck Norris says "Drill now in America for oil" Message-ID: <38KPUbcQPgV4@eisner.encompasserve.org> In article <820d9f14-9ab1-4120-8e46-8911cd92dbed@r66g2000hsg.googlegroups.com>, ultradwc@gmail.com writes: > an interview on fox You get your information from Fox and you listen to an actor for environmental advice? Q.E.D. ------------------------------ Date: 13 Jun 2008 20:50:17 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Chuck Norris says "Drill now in America for oil" Message-ID: <4852dd89$0$11640$607ed4bc@cv.net> In article <4852b42a$0$12336$c3e8da3@news.astraweb.com>, JF Mezei writes: >VAXman- @SendSpamHere.ORG wrote: > >> We have more than a few days worth of oil -- don't fool yourself. However, >> the problem isn't the lack of supply. We've actually been using less oil >> than in the past. What is needed is some constraints on the speculation >> that has been driving the prices to levels way, way outside of the supply >> and demand economic equilibrium point. > >The current regime in Washington has taken every step to help its oil >business. Heck, at the current oil prices, even Bush even might be able >to run his oil company without running it into thr ground and requiring >a bail out by a certain arab family. JF, reality time. Stop listening to the US hate-bias media you have been and start learning some reality. >If the USA government wanted to send a strong message about future oil >consumption, it would simply announce that *all* passenger vehicles >would have to abide by strong fuel consumption rules. (right now, they >apply to "cars", and those contraptions called SUVs were lobbied by the >car industry as "trucks" and not bound by those standards). I don't like SUVs any more than you do. However, the US oil consumption figures are down from last year and the prices keep going up. I, for one, do not drive to work. I conserve more fuel than the average American and, I'd wager, the average Canadian. >As long as the USA refuses to move towards seriously reducing oil >consumption (as a nation), the USA (as a nation) cannot tell China and >India to move in the same direction. As a result, worldwide demand >continues to grow, and production isn't going to increase much. So >prices go up. > >The USA addiction to oil is far more serious than americans realise. The >money is shifting to new countries like the middle east, singapore and >china. And as the USD drops in value, USA corporations become apetizing. >(One reason InBev wants to buy the owner of the Budweiser brand.) Look >at where the money came to bail out a number of USA banks recently: >middle east (a UAE fund now owns a big chunk of Citibank), Singapore and >China. > >A few years ago, democrats went ballistic when the DPW transction was >about to be signed. God forbid a foreign corporation would lease space >in USA ports. Well those dems you keep harping about haven't done a whole hell of a lot while they've been in power in congress. It's time they all stop porking out and start doing something other than finding new ways to grow tax-fed bureaucracies. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" http://tmesis.com/drat.html ------------------------------ Date: Fri, 13 Jun 2008 17:26:03 -0400 From: JF Mezei Subject: Re: Chuck Norris says "Drill now in America for oil" Message-ID: <4852e70f$0$12311$c3e8da3@news.astraweb.com> VAXman- @SendSpamHere.ORG wrote: > JF, reality time. Stop listening to the US hate-bias media you have been > and start learning some reality. Are you guys aware that Bush passed an executive order granting immunity to his croonies (namely Cheney, Rumsfeld, Wolfowitz) to protect them against the supreme court decision that ruled that they had condoned torture ? It wasn't long ago that the USA was ready to impeach a president for having lyed about a blow job. But performing torture, suspending habeas corpus, suspending human rights seems pretty acceptable to americans. Why ? You guys complain about Iraqi government not being functional. Do americans know that the USA government is preventing the iraqi government from being functional ? Do americans realise that all the money being spent in Iraq goes to USA firms instead of hiring iraqi companies to rebuild the country and reduce its unemployment ? (hint: wqhen employment si low, there is less crime because people are happier). Are you aware that Bush and his croonies have imposed conditions to all the money they give towards AIDS in africa ? These conditions prevent all aid workers from dealing with prostitutes and drug users. These are the very people who need to be educated on AIDS. Get a prostitute to force her customers to wear conduoms and you've just save a whole bunch of men from getting Aids. (guess who spreads aids in africa ? It s the prostitutes and these are the very people who need to be involved, the very people Bush as targetted to be isolated from any work done by the NGOs that are partly funded by the USA. The USA's refusal to participate in environmental issues, its refusal to force higher fuel standards and begin to wein people from their cars is in part responsible for the high fuel prices. India and China simply point to the USA as the precedent that allows them to grow their fuel use without restrictions. And that impacts the rest of the world. There was a period where the rise in the oil price didn't affect the rest of the world because it was simply to compensate for the drop in the value of the USA. But now, the increases affect everyone and that is a worldwide issue. > I don't like SUVs any more than you do. However, the US oil consumption > figures are down from last year and the prices keep going up. Woopty do. You are still, by far, the largest per capita consumer of oil. (Note that Canada, under George W Harper's regime has followed Bush's mislead and we are the worse performer for Kyoto) ------------------------------ Date: Fri, 13 Jun 2008 22:25:59 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: Chuck Norris says "Drill now in America for oil" Message-ID: In article <4852e70f$0$12311$c3e8da3@news.astraweb.com>, JF Mezei writes: > It wasn't long ago that the USA was ready to impeach a president for > having lyed about a blow job. But performing torture, suspending habeas > corpus, suspending human rights seems pretty acceptable to americans. No surprise here. Consider what brutality you can see in a PG film and how little sex it takes to get an R or NC-17. (Also interesting: check what films have what ratings in other countries to get an idea what a given culture deems dangerous to youth.) ------------------------------ Date: Fri, 13 Jun 2008 20:39:36 -0500 From: David J Dachtera Subject: Re: Chuck Norris says "Drill now in America for oil" Message-ID: <48532158.662B3C66@spam.comcast.net> ultradwc@gmail.com wrote: > > we have 68 billion barrels of oil sitting in this country but the > democrats > who are in bed with the environmentalists will not allow drilling ... > > well, my family and this country is more important than snails or > owls, and Chcuk Norris agrees ... > > http://youtube.com/watch?v=cXOdI5XzcM0&feature=related One thing ya gotta say for Boob - he knows how to up the post count for this ng! D.J.D. ------------------------------ Date: Fri, 13 Jun 2008 18:33:44 -0400 From: "John Smith" Subject: Does anyone know if and how VMS figured in this? Message-ID: <17912$4852f171$4c0aab67$21546@TEKSAVVY.COM> http://www.pcworld.com/businesscenter/article/147054/stock_exchange_risks_sanctions_over_system_outages.html Stock Exchange Risks Sanctions Over System Outages Mikael Ricknäs, IDG News Service Friday, June 13, 2008 7:50 AM PDT After two trade outages, financial regulators in Sweden and Norway are taking a close look at stock exchanges operated by Nasdaq OMX in several Nordic business hubs, and could impose fines if the company is found to be negligent. On the morning of June 2 trading did not open until 9:40 a.m. on the OMX Nordic Exchange and the Oslo Stock Exchange. On June 3 the same problem recurred, but trading didn't start until 2:30 p.m. Stock exchanges in Stockholm, Copenhagen, Helsinki, Reykjavik and Oslo were all affected. "It was a series of events -- related to our operating system, network and the new version of our trading system Saxess, that collided, and resulted in the two outages," said Carl Norell, press officer at Nasdaq OMX Group, which owns OMX Nordic Exchange. On Wednesday it turned over a report detailing the events that led up the trade outages to Finansinspektionen, the financial supervisory authority in Sweden. Its investigation, expected to take about a month, will see if Nasdaq OMX handled everything by regulation, according to Annika Von Haartman, who is in charge of market oversight at Finansinspektionen. If it concludes that Nasdaq OMX somehow failed, it has a number sanction options. "We can give it a reprimand, have it pay a fine between 5,000 (US$820) and 50 million Swedish kronor or as a last resort revoke its license," said Von Haartman. She doesn't want to speculate on what a proportional sanction would be, but underscores that Finansinspektionen takes this matter very seriously. The Norwegian financial supervisory authority Kredittilsynet is waiting for a full report from the Oslo Stock Exchange, so it can launch its investigation into the outages, according to special advisor Knut Godager. Oslo Stock Exchange (Oslo Børs) isn't a part of OMX Nordic Exchange, but uses the same trading system. "We will turn in full a report by the end of next week," said Tor Arne Olsen, spokesman at Oslo Stock Exchange. At the beginning of next week it receive a more detailed root cause analysis from Nasdaq OMX, according to Norell. Kredittilsynet can also dole out a reprimand, or demand system improvements, but these things usually solve themselves, since it's also in the interest of the stock exchange that its systems work properly, according to Godager. Finally there remains the question of whether the outages indicate a breach of the service level agreement between Oslo Stock Exchange and OMX Technology. Olsen doesn't believe so, but since uptime is measured over a longer period that remains to seen. "We still have every confidence in the platform," said Olsen. ------------------------------ Date: Fri, 13 Jun 2008 18:55:07 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: http://vmspython.dyndns.org/DownloadAndInstallationPython Message-ID: OK, it seems to be working OK. (The compile is taking a while even on an ALPHAserver 1200.) However, consider the following text: $ unzip jfppy0010.zip $! which extract jfppy0010.dsk $ ld CONNECT jfppy0010.dsk lda2: ! <- Don't forget to give a device and directory for the .DSK $ mount/system lda2: jfppy0010 $ @lda2:[000000]python_startup $! this defines system logicals names However, the files are called JFPLIB0002.DSK;1 JFPPY0011.DSK;1 That is, the files at the links at the top of the web page pointed at by the URL and where one can read the above commands. I made the obvious change, but the text needs to be corrected. ------------------------------ Date: Fri, 13 Jun 2008 18:59:13 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: http://vmspython.dyndns.org/DownloadAndInstallationPython Message-ID: In article , helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: > OK, it seems to be working OK. (The compile is taking a while even on > an ALPHAserver 1200.) Doesn't work cleanly out of the box: Sorry [Errno 5] i/o error: '/python_root/lib/SimpleHTTPServer.pyo' Compiling /python_root/lib/SimpleXMLRPCServer.py ... Sorry [Errno 5] i/o error: '/python_root/lib/SimpleXMLRPCServer.pyo' Compiling /python_root/lib/SocketServer.py ... Sorry [Errno 5] i/o error: '/python_root/lib/SocketServer.pyo' Hopefully this will be OK for the limited use I need to put python to now. ------------------------------ Date: Fri, 13 Jun 2008 19:00:17 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: http://vmspython.dyndns.org/DownloadAndInstallationPython Message-ID: In article , helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: > In article , helbig@astro.multiCLOTHESvax.de > (Phillip Helbig---remove CLOTHES to reply) writes: > > > OK, it seems to be working OK. (The compile is taking a while even on > > an ALPHAserver 1200.) > > Doesn't work cleanly out of the box: > > Sorry [Errno 5] i/o error: '/python_root/lib/SimpleHTTPServer.pyo' > Compiling /python_root/lib/SimpleXMLRPCServer.py ... > Sorry [Errno 5] i/o error: '/python_root/lib/SimpleXMLRPCServer.pyo' > Compiling /python_root/lib/SocketServer.py ... > Sorry [Errno 5] i/o error: '/python_root/lib/SocketServer.pyo' Those were just a few examples; there are quite a few more such errors. ------------------------------ Date: Fri, 13 Jun 2008 22:37:14 +0200 From: =?ISO-8859-1?Q?Jean-Fran=E7ois_Pi=E9ronne?= Subject: Re: http://vmspython.dyndns.org/DownloadAndInstallationPython Message-ID: <4852da7a$0$19724$426a74cc@news.free.fr> Phillip Helbig---remove CLOTHES to reply wrote: > In article , helbig@astro.multiCLOTHESvax.de > (Phillip Helbig---remove CLOTHES to reply) writes: > >> In article , helbig@astro.multiCLOTHESvax.de >> (Phillip Helbig---remove CLOTHES to reply) writes: >> >>> OK, it seems to be working OK. (The compile is taking a while even on >>> an ALPHAserver 1200.) >> Doesn't work cleanly out of the box: >> >> Sorry [Errno 5] i/o error: '/python_root/lib/SimpleHTTPServer.pyo' >> Compiling /python_root/lib/SimpleXMLRPCServer.py ... >> Sorry [Errno 5] i/o error: '/python_root/lib/SimpleXMLRPCServer.pyo' >> Compiling /python_root/lib/SocketServer.py ... >> Sorry [Errno 5] i/o error: '/python_root/lib/SocketServer.pyo' > > Those were just a few examples; there are quite a few more such errors. > This is fixed in latest version (jfppy0011), donwload an installed the latest version. JFP ------------------------------ Date: Fri, 13 Jun 2008 22:23:03 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: http://vmspython.dyndns.org/DownloadAndInstallationPython Message-ID: In article <4852da7a$0$19724$426a74cc@news.free.fr>, =?ISO-8859-1?Q?Jean-Fran=E7ois_Pi=E9ronne?= writes: > This is fixed in latest version (jfppy0011), donwload an installed the > latest version. I'm pretty sure that's what I have. I downloaded it yesterday; has anything changed since then? ------------------------------ Date: 13 Jun 2008 13:44:53 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Interesting job ad from HP Message-ID: On Jun 13, 3:58=A0am, Michael Kraemer wrote: > I doubt that you have seen many production environments > or that you have ever seen more than one HP-UX installation. > We use various flavours Unix for about just everything > for many years, including =A0mission critical real-time apps > (is radiotherapy critical enough for you ?), > and the uptime is only limited by the availability of power. That meaning of realtime is something almost any system can handle. The amount of radiation used varies very little if the computer gets a few milliseconds behind. The only real reliability requirement is that something turn off the radiation if the computer fails. Hard real time means if your behind by 1 millisecond you've failed. Not you're good enough, not you almost got it, you failed. If that doesn't describe your system, we're not impressed. > I have the impression we must live on different planets. > Again, this is plain nonsense. > I use both, Google and eBay many times per day (and sometimes > even per night). Availability over the past 3 or 4 years: 100%. > May be you should check your ISP. Google and eBay can survive multiple system outages if the end user gets his data. Google and eBay will make just as much money as long as they process transactions at worst fast enough to keep the users happy and 24 hours worth every 24 hours. Some systems can't. Some systems loose large sums of money for every minute each computer is down and aren't so readily implemented over large numbers of redundant computers. If that doesn't describe your system, we're not impressed. ------------------------------ Date: Fri, 13 Jun 2008 15:28:19 -0400 From: JF Mezei Subject: Re: Interesting job ad from HP Message-ID: <4852cbbd$0$7275$c3e8da3@news.astraweb.com> Bob Koehler wrote: > That meaning of realtime is something almost any system can handle. > The amount of radiation used varies very little if the computer > gets a few milliseconds behind. With advances in technooogy, I wonder how much still *really* needs true real-time operating system. Consider the USA segment on the space station. It uses a military equivalent of token passing slow network for the critical systems. The thinking was that it provided predictable response time. But when you consider today's ethernet, the variations in response time are probably well below the actual time this military system has for its normal operations. (aka: the ethernet would still be faster than the military system even during bad times for the ethernet segment). The space station itself, as a vehicle, does not need "real time" because even if everything fails, it continues in the same direction for months. But the shuttle is a different story. Its computers date back from the 1970s, and at that time, speed and memory constraints required very smart coding design and true real time OS. But today, with 4ghz PCs, I would venture to guess that even Windows might be able to handle the job, provided the application software was well written and that you didn't play space invadors game at the same time as the shuttle was launching. One also needs to look at how the "orders" are distributed. Does the computer tell the robot to start motor X in a direction, and X milliseconds later, tells it to stop motor X. Or does it give the robot an order move motor X by Y steps/clicks ? If the order is in "steps", then the need for absolutely precise timing between computer and robot is not as big. > Google and eBay can survive multiple system outages if the end user > gets his data. Actually, I thinnk it is far worse than that. Users won't really notice if Google, due to failure of a database, only returns half of the results when you search for the word "chocolate". As long as it returns any results, usres think it is working. And as long as it includes ads, Google probably considers that transaction to be 100% succesful. There is no way a user can know if Google has truily returned all posible results or not. This is very different from an industrial application where if the hydraulic press is ordered to come down, but the order to remove the robot arm that placed an object under the press was late, the hydraulic press will end up crushing an important production robot along with the part. That tends to force the whole production line to stop because it was broken. ------------------------------ Date: Fri, 13 Jun 2008 21:11:32 GMT From: Roger Ivie Subject: Re: Interesting job ad from HP Message-ID: On 2008-06-13, JF Mezei wrote: > With advances in technooogy, I wonder how much still *really* needs true > real-time operating system. Well, I can't give any real examples lately (I've not investigated the behavior of general-purpose OSes recently; we just started out by specing vxWorks on my most recent project), but I can give you an example of the sort of things that can happen from a while ago. Back in the OSF/1 V1.3 days, I was involved in developing hardware that would allow the interrupt latency to be determined and was also involved in writing device drivers, specifically an 8-port TURBOchannel device. I kept having trouble losing data, so I did some investigation. Turned out that when a process exited, OSF/1 disabled interrupts while tearing down the process. The amount of time for which interrupts was disabled was proportional to the amount of memory allocated to the job. If a job was sufficiently large (IIRC, had allocated 150MB) interrupts would be disabled so long that the eight-byte FIFO in our serial port would lose data at *300* baud! Another annoying thing you could do to cause long interrupt latency on OSF/1 and MIPS Ultrix was grab a window with the mouse and move it around. In contrast, the same Alpha hardware under VMS consistently delivered interrupt latencies in the 15 microsecond range, regardless of load. Now, it's been a long time, machines have gotten faster, but I have seen nothing to give me warm fuzzies that interrupt latency in general-purpose systems has gotten more predictable. But I haven't measured anything recently, so I can't say for certain. Also bear in mind that a fast processor doesn't mean fast response to external stimuli. In fact, in at least one case I know it doesn't: the interrupt latency in the VAXstation 4000/60 was *much* better than the latency in the 4000/90. These days, device registers are so far away from the processor that I'm getting worried about it. My most recent project involved a VMEbus SBC from Motorola. You'd expect that to be nice and tight, but the VMEbus is behind a PCI bridge and a PCI to VMEbus adapter. I don't know that it's even possible to buy "native" VMEbus machines anymore. As the processors get faster, it gets more expensive for them to get to the hardware, in terms of the effect of the I/O transfer on the processor's performance. The processors are made faster by moving things closer to them, caching, buffering, speculative execution, etc; all of those get blown away when you need to do something as simple as read a register sitting in an I/O device. I've not yet figured out a way to measure interrupt latency on our new Itanium machine at work, but I'm thinking about it. An Itanium processor has a lot of state to save, so I don't expect it to provide stellar interrupt latency. I also need to figure out how to measure the latency between the processor and our I/O devices; in our case, our most interesting devices are sitting on a VMEbus on the end of a cable attached to a PCI adapter which is behind a PCI bridge hanging off a PCI interface to the processor bus. I'm suprised that *any* I/O gets done! > Does the computer tell the robot to start motor X in a direction, and X > milliseconds later, tells it to stop motor X. Or does it give the robot > an order move motor X by Y steps/clicks ? If the order is in "steps", > then the need for absolutely precise timing between computer and robot > is not as big. This just moves the real-time problem from "the computer" to the embedded system in the robot (which is probably also a computer). *Someone* has to meter out the steps. And that someone has to meet the timing requirements. -- roger ivie rivie@ridgenet.net ------------------------------ Date: 13 Jun 2008 16:20:23 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Interesting job ad from HP Message-ID: <5qiyDRDtiQgD@eisner.encompasserve.org> In article <4852cbbd$0$7275$c3e8da3@news.astraweb.com>, JF Mezei writes: > Bob Koehler wrote: > >> That meaning of realtime is something almost any system can handle. >> The amount of radiation used varies very little if the computer >> gets a few milliseconds behind. > > With advances in technooogy, I wonder how much still *really* needs true > real-time operating system. > Speed is not the same as deterministic. I have a 12MHz 8085 that does 25ms work every 23.5 ms. It has done this for over 20 years without a hitch. I don't care how fast the CPU is as long as it makes it before 25 ms. I have a 40MIP RISC CPU that can't make 100ms for 24 hours straight because the UNIX kernel on it might decide to do something else for 500ms. That RISC CPU could eat floating point circles around the 8085's integer pace, but the slower system is the one that can do its job. And the 25ms requirement comes from hardware that would cost so much to replace it actually would take an Act of Congress. And we had a PDP-11/70 running RSX-11M, a VAX 11/785 running VMS, and now a 12MIP VAX 4000 all do that same application as that RISC/UNIX box without a blip. For over 3 decades starting with the 11/70. > The space station itself, as a vehicle, does not need "real time" > because even if everything fails, it continues in the same direction for > months. And the station cannot maintain attitude, so power is lost and the astronauts freeze to death. Well, no, there are backup systems to protect human life. But real-time is not as important on that part of the system as reliability. The hard real-time requirements are elsewhere in the station. > But the shuttle is a different story. Its computers date back from the > 1970s, and at that time, speed and memory constraints required very > smart coding design and true real time OS. But today, with 4ghz PCs, I > would venture to guess that even Windows might be able to handle the > job, provided the application software was well written and that you > didn't play space invadors game at the same time as the shuttle was > launching. You'ld guess wrong. You're looking at the wrong issue. And your information is way out of date. But during re-entry there isn't time for that BSOD that was sitting at the desk next to mine the other morning. And hard real-time on Windows will make a tradtional timesharing UNIX kernel look good. I've got a system in the next room that would have been easy to do on a hard real-time kernel. Even Solaris' real-time extenstions would probably be enough. But it's on Windows and it's having a hard time doing things once a second with its processor running over 2GHz. The kinds of things I did for years on a VAX 11/780 with no trouble. > One also needs to look at how the "orders" are distributed. > > Does the computer tell the robot to start motor X in a direction, and X > milliseconds later, tells it to stop motor X. Or does it give the robot > an order move motor X by Y steps/clicks ? If the order is in "steps", > then the need for absolutely precise timing between computer and robot > is not as big. That depends on the hardware design. If someone gives you $15 million hardware designed in the former manner because the latter would have cost $20 million, you'ld better pick a system that can handle it. ------------------------------ Date: Fri, 13 Jun 2008 23:45:36 +0200 From: Michael Kraemer Subject: Re: Interesting job ad from HP Message-ID: JF Mezei schrieb: > > With advances in technooogy, I wonder how much still *really* needs true > real-time operating system. > > Consider the USA segment on the space station. It uses a military > equivalent of token passing slow network for the critical systems. The > thinking was that it provided predictable response time. But when you > consider today's ethernet, the variations in response time are probably > well below the actual time this military system has for its normal > operations. (aka: the ethernet would still be faster than the military > system even during bad times for the ethernet segment). You probably know that it's not enough for a real-time system to be just fast. The real requirement is deterministic behaviour, i.e. a guaranteed response time. There might be applications where one might get away with a faster CPU, but if one needs 100% deterministic behaviour, this is not enough. Ethernet is not deterministic, nor is Windoze, let alone a combination of both. Been there, tried it, failed. We had a setup where Windoze+sockets had to be used because the necessary off-the-shelf software (LabView) is/was only available on BillyBoxes. The problem could be solved by going back to old-fashioned VME and a classical real-time OS, where 25 MHz CPUs sometimes are enough. ------------------------------ Date: Fri, 13 Jun 2008 23:58:23 +0200 From: Michael Kraemer Subject: Re: Interesting job ad from HP Message-ID: Bob Koehler schrieb: > > That meaning of realtime is something almost any system can handle. > The amount of radiation used varies very little if the computer > gets a few milliseconds behind. The only real reliability > requirement is that something turn off the radiation if the computer > fails. This is not true for modern technologies like IMRT and even more so scanned charged particles. They need permanent control of the partial beams as far as intensity, position and shape are concerned. This can go way down to the millisecond level. But that was not really the point. The OP claimed that Unix systems can't stay up for a few days or weeks (there was no mention of realtime), which is, of course, nonsense. > Google and eBay can survive multiple system outages if the end user > gets his data. Google and eBay will make just as much money as long > as they process transactions at worst fast enough to keep the users > happy and 24 hours worth every 24 hours. Some systems can't. Some > systems loose large sums of money for every minute each computer > is down and aren't so readily implemented over large numbers of > redundant computers. If that doesn't describe your system, we're > not impressed. I couldn't care less if you're impressed or not. I simply counter some of the unsubstantiated claims made here. One of these was that a *x system can't provide 100% service to end users. eBay and Google are counter examples that this is not true, and as far as I know they run on Linux. ------------------------------ Date: Fri, 13 Jun 2008 18:05:15 -0400 From: JF Mezei Subject: Re: Interesting job ad from HP Message-ID: <4852f0f4$0$31243$c3e8da3@news.astraweb.com> Michael Kraemer wrote: > to be just fast. The real requirement is deterministic behaviour, > i.e. a guaranteed response time. There might be applications > where one might get away with a faster CPU, but if one needs > 100% deterministic behaviour, this is not enough. > Ethernet is not deterministic, nor is Windoze, If the requirements aere to be able to transmit a frame within say 50ms, and ethernet varies between 1 and 40ms, then ethernet would still be well within the performance norms. However, software would have to be written accordingly. (and packet from device 2 might arrive before packet from device 1 once, and vice versa the next time, you just need to architect the system (overall) to be aware of this. If I take an off-the-shelf OS-X unix, it is laden with scheduled utilities (for instance, to rebuild the spotlight search index of all your documents). Obviously, one would have to strip it down to a bare bones OS to not have any of those pesky jobs start and slow the system down. Same with Windows. If you remove all the crud that runs in foreground and background of windows, wouldn't its kernel behave properly ? Isn't the kernel fairly similar to VMS in that a process with priority higher than a certain number doesn't get interrupted ? ------------------------------ Date: Sat, 14 Jun 2008 00:06:23 +0200 From: Michael Kraemer Subject: Re: Interesting job ad from HP Message-ID: Roger Ivie schrieb: > > Now, it's been a long time, machines have gotten faster, but I have seen > nothing to give me warm fuzzies that interrupt latency in > general-purpose systems has gotten more predictable. Well, that's why real-time OSs are still around :-) > These days, device registers are so far away from the processor that I'm > getting worried about it. My most recent project involved a VMEbus > SBC from Motorola. You'd expect that to be nice and tight, but the VMEbus > is behind a PCI bridge and a PCI to VMEbus adapter. I don't know that > it's even possible to buy "native" VMEbus machines anymore. You can. Although the leader in this business, Motorola, has thrown the towel, as they did with their semiconductors. They concentrate on their core compentency (ringtones). 68k is a bit out-of-fashion these days, but you can buy e.g. PowerPC based VME modules. ------------------------------ Date: Fri, 13 Jun 2008 22:20:15 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: Interesting job ad from HP Message-ID: In article <4852cbbd$0$7275$c3e8da3@news.astraweb.com>, JF Mezei writes: > But the shuttle is a different story. Its computers date back from the > 1970s, and at that time, speed and memory constraints required very > smart coding design and true real time OS. At least the original shuttle computers (I believe there were 4, all calculating the same stuff, and things were "go" if 3 of the 4 agreed, or something like that---and a fifth backup unit) IIRC had core memory, i.e. the little magnetic rings threaded on wires. ------------------------------ Date: Sat, 14 Jun 2008 00:30:34 +0200 From: Michael Kraemer Subject: Re: Interesting job ad from HP Message-ID: JF Mezei schrieb: > If the requirements aere to be able to transmit a frame within say 50ms, > and ethernet varies between 1 and 40ms, then ethernet would still be > well within the performance norms. well, as said, depends on application. If requirements say 40ms, than things might go well most of the time, but sometimes not. Such a situation is worse than a simple "clean" crash. In our case the delay sometimes was extremely long due to Windoze doing something secretely in the background. It could not be isolated (of course, we're talking crapware here). > However, software would have to be written accordingly. (and packet from > device 2 might arrive before packet from device 1 once, and vice versa > the next time, you just need to architect the system (overall) to be > aware of this. Well, in reality, software is only part of the game. At some point, one has to get an apparatus up and running. A clean dedicated RTOS (such as Lynx, OS-9 and the others mentioned here) is a safer bet than a tweaked general purpose OS. > If I take an off-the-shelf OS-X unix, it is laden with scheduled > utilities (for instance, to rebuild the spotlight search index of all > your documents). Obviously, one would have to strip it down to a bare > bones OS to not have any of those pesky jobs start and slow the system down. > > Same with Windows. If you remove all the crud that runs in foreground > and background of windows, wouldn't its kernel behave properly ? Isn't > the kernel fairly similar to VMS in that a process with priority higher > than a certain number doesn't get interrupted ? I don't know. Probably unfeasible as far as Windoze is concerned. And somehow a moot point as long as we can have dedicated RTOSs. ------------------------------ Date: Fri, 13 Jun 2008 23:40:01 GMT From: Roger Ivie Subject: Re: Interesting job ad from HP Message-ID: On 2008-06-13, Michael Kraemer wrote: > Roger Ivie schrieb: >> I don't know that >> it's even possible to buy "native" VMEbus machines anymore. > > You can. Although the leader in this business, Motorola, > has thrown the towel, as they did with their semiconductors. > They concentrate on their core compentency (ringtones). > 68k is a bit out-of-fashion these days, > but you can buy e.g. PowerPC based VME modules. Well, I have a PowerPC based VME module, it's just not what I consider "native". It's an MVME5500. Here's how it works: PowerPC -> host bridge -> PCI -> PCI Bridge -> Universe -> VMEbus I would prefer is something more like: PowerPC -> VMEbus. Basically, the MVME5500 is a PCI-based PowerPC plopped on the same board with a PCI to VMEbus adapter. Not what I would consider "native" at all. -- roger ivie rivie@ridgenet.net ------------------------------ Date: 13 Jun 2008 13:57:10 -0500 From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) Subject: OpenVMS Alpha 8.4 release date and new functionality ? Message-ID: Does anyone have an idea of when OpenVMS 8.4 Alpha is likely to ship and what kind of new functionality is likely to be in it ? I've found the slide on the Roadmap, but it's very vague on specifics and I couldn't find anything else that addressed this. Thanks, Simon. -- Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP Microsoft: Bringing you 1980's technology to a 21st century world ------------------------------ Date: Fri, 13 Jun 2008 18:59:37 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: OpenVMS Alpha 8.4 release date and new functionality ? Message-ID: In article , clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) writes: > Does anyone have an idea of when OpenVMS 8.4 Alpha is likely to ship > and what kind of new functionality is likely to be in it ? > > I've found the slide on the Roadmap, but it's very vague on specifics > and I couldn't find anything else that addressed this. Also: What is the latest version of VMS for VAX it will cluster with. ------------------------------ Date: Fri, 13 Jun 2008 15:31:05 -0400 From: JF Mezei Subject: Re: OpenVMS Alpha 8.4 release date and new functionality ? Message-ID: <4852cc63$0$7275$c3e8da3@news.astraweb.com> Simon Clubley wrote: > Does anyone have an idea of when OpenVMS 8.4 Alpha is likely to ship > and what kind of new functionality is likely to be in it ? If this was discussed during the bootcamp, it means that all of the information about 8.4 is now under an NDA and nobody can discuss it :- :-) :-) :-) ;-) :-) :-) :-) It will come when it will come. ------------------------------ Date: Fri, 13 Jun 2008 20:47:42 -0500 From: David J Dachtera Subject: Re: OpenVMS for Small and Medium Businesses? Message-ID: <4853233E.BCCA9ADC@spam.comcast.net> D Gillbilly wrote: > > OpenVMS for Small and Medium Businesses? > > http://64.223.189.234/node/719 > > Good article pointing out previous Vendor neglect. The original version of this page dates back to Jan of 1998: http://www.djesys.com/vms/soho/ D.J.D. ------------------------------ Date: Fri, 13 Jun 2008 14:19:37 -0400 From: JF Mezei Subject: Re: PCL and PDF support in DCPS. Message-ID: <4852bc0f$0$31210$c3e8da3@news.astraweb.com> Paul Anderson wrote: > And we'd be happy to hear comments about how the PDF printing should be > implemented. The feature has been on our to-do list for quite a while > but is not scheduled for any future release. Perhaps a concept of "plug in" for DCPS. Make the architecture possible for anyone to write code that is used by DCPS when printing certain kinds of documents. You feed a .PDF file ? It calls the code defined to handle .PDF You feed a .JPG file ? It calls the code defined to handle .JPG (aka: convert it to postscript). Something akin to the CDA converter library where there was a defined API allowing people to write converters. But done in a printing context. ------------------------------ Date: Fri, 13 Jun 2008 22:43:42 +0200 From: =?ISO-8859-1?Q?Jean-Fran=E7ois_Pi=E9ronne?= Subject: Re: Python for VMS Message-ID: <4852dbfe$0$27506$426a74cc@news.free.fr> R.A.Omond wrote: [snip] > > I second this; it's very easy to setup. There's no VAX > version that I am aware of (possibly because of lack of > IEEE floating-point ?). I've got it setup using logical [snip] That's correct, Python need IEEE floating-point, 64 bits integer support ('long long int' C type) and ODS-5. JFP ------------------------------ Date: Fri, 13 Jun 2008 20:40:48 -0500 From: David J Dachtera Subject: Re: TCPware 5.8 and MIME Message-ID: <485321A0.12EB6C5E@spam.comcast.net> Neil Rieck wrote: > > On Jun 12, 3:21 pm, Neil Rieck wrote: > > Folks, > > > > I just installed TCPware 5.8 and discovered that the good folks at PSC > > have finally fixed the MIME attachment problem. > > > > $!============================================================ > > $! title : mime_hack.com > > $! author : Neil Rieck > > $! created: 2008-06-12 > > $!============================================================ > > $! > > $ say := write sys$output > > $! > > $! method #1 (works as-is with TCPware 5.8) > > $! > > $ say "mailing test #1" > > $ mail/subj="mime test 1"/for/type=1 SCR-2008H1.zip > > "neil.ri...@bell.ca" > > $! > > $ say "building MIME file" > > $ wait 0:00:01 > > $ MIME ! start the MIME application > > new/noedit neil.mime ! create a new mime document > > add/encod=base64 SCR-2008H1.zip ! > > save ! save mime document > > exit ! prompt (leave mime) > > $! > > $! method #2 (works with new TCPware 5.8 logical) > > $! > > $ say "mailing test #2" > > $ def tcpware_smtp_allow_mime_send y ! we want new functionality > > $ mail/subj="mime test 2" neil.mime "neil.ri...@bell.ca" > > $ say "adios" > > > > Neil Rieck > > Waterloo, Ontario, Canada.http://www3.sympatico.ca/n.rieck/ > > The previous example assumes that this definition has been made > somewhere else (like SYS$MANAGER:SYLOGIN.COM) > > $MIME :== $SYS$SYSTEM:MIME.EXE > > NSR ...or let SYS$SYSTEM: be an element of your DCL$PATH search list (*CAUTION*!!!) D.J.D. ------------------------------ Date: 13 Jun 2008 14:16:55 -0500 From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) Subject: What happened to VAXeln after DEC stopped selling it ? Message-ID: Does anyone know what happened to VAXeln ? Was it released into the community, just discontinued, or sold on to a third party ? The reason I was asking is that I am curious to find out more about it. I've found the Bookreader type manuals at http://www.sysworks.com.au/swadm_dat_root/cddoc04jan1/VAXELN_41.html and http://www.sysworks.com.au/swadm_dat_root/vaxdocdec96/d33vza11.html Does anyone know if the manuals are available online in a more friendly format ? In case anyone's curious, this is just for personal interest; I would like to know more about it's internal design. Thanks, Simon. -- Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP Microsoft: Bringing you 1980's technology to a 21st century world ------------------------------ Date: Fri, 13 Jun 2008 15:40:42 -0400 From: JF Mezei Subject: Re: What happened to VAXeln after DEC stopped selling it ? Message-ID: <4852ce5d$0$12277$c3e8da3@news.astraweb.com> Simon Clubley wrote: > Does anyone know what happened to VAXeln ? Wasn't ELN used in DECservers ? ------------------------------ Date: Fri, 13 Jun 2008 16:13:04 -0500 (CDT) From: sms@antinode.info (Steven M. Schweda) Subject: Re: What happened to VAXeln after DEC stopped selling it ? Message-ID: <08061316130437_20200492@antinode.info> From: JF Mezei > Wasn't ELN used in DECservers ? I think that I remember a VAXstation-based X terminal using VAXELN. No bets. ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-info 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: 13 Jun 2008 20:53:30 GMT From: VAXman- @SendSpamHere.ORG Subject: What is a Weendoze .CHM file? Message-ID: <4852de4a$0$11640$607ed4bc@cv.net> I have a new piece of hardware and the user manual is on CD in about 10 different languages. In the English folder is a file called MANUAL.CHM What do I use to read it? -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" http://tmesis.com/drat.html ------------------------------ Date: Fri, 13 Jun 2008 22:47:16 +0100 From: "Robert Jarratt" Subject: Re: What is a Weendoze .CHM file? Message-ID: wrote in message news:4852de4a$0$11640$607ed4bc@cv.net... >I have a new piece of hardware and the user manual is on CD in about 10 > different languages. In the English folder is a file called MANUAL.CHM > What do I use to read it? > > -- > VAXman- A Bored Certified VMS Kernel Mode Hacker > VAXman(at)TMESIS(dot)COM > > "Well my son, life is like a beanstalk, isn't it?" > > http://tmesis.com/drat.html It is a Compiled HTML help file. I think any reasonably modern version of Windows will include a reader for it. Wikipedia has some information here: http://en.wikipedia.org/wiki/Microsoft_Compiled_HTML_Help Regards Rob ------------------------------ Date: 13 Jun 2008 16:59:40 -0500 From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) Subject: Re: What is a Weendoze .CHM file? Message-ID: <1RyTnZiijmjb@eisner.encompasserve.org> In article , "Robert Jarratt" writes: > > wrote in message > news:4852de4a$0$11640$607ed4bc@cv.net... >>I have a new piece of hardware and the user manual is on CD in about 10 >> different languages. In the English folder is a file called MANUAL.CHM >> What do I use to read it? >> > It is a Compiled HTML help file. I think any reasonably modern version of > Windows will include a reader for it. Wikipedia has some information here: > http://en.wikipedia.org/wiki/Microsoft_Compiled_HTML_Help > There's also a program, xchm, which I've built and used under Linux in the past. Note that xchm is a frontend to a library called chmlib. Simon. -- Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP Microsoft: Bringing you 1980's technology to a 21st century world ------------------------------ Date: 14 Jun 2008 00:02:41 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: What is a Weendoze .CHM file? Message-ID: <48530aa1$0$5010$607ed4bc@cv.net> In article , "Robert Jarratt" writes: > > wrote in message >news:4852de4a$0$11640$607ed4bc@cv.net... >>I have a new piece of hardware and the user manual is on CD in about 10 >> different languages. In the English folder is a file called MANUAL.CHM >> What do I use to read it? >> >> -- >> VAXman- A Bored Certified VMS Kernel Mode Hacker >> VAXman(at)TMESIS(dot)COM >> >> "Well my son, life is like a beanstalk, isn't it?" >> >> http://tmesis.com/drat.html > >It is a Compiled HTML help file. I think any reasonably modern version of >Windows will include a reader for it. Wikipedia has some information here: >http://en.wikipedia.org/wiki/Microsoft_Compiled_HTML_Help ...and therein lies the problem. No Weendoze. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" http://tmesis.com/drat.html ------------------------------ Date: Sat, 14 Jun 2008 02:15:34 +0200 From: Michael Kraemer Subject: Re: What is a Weendoze .CHM file? Message-ID: Robert Jarratt schrieb: > > It is a Compiled HTML help file. I think any reasonably modern version of > Windows will include a reader for it. Wikipedia has some information here: > http://en.wikipedia.org/wiki/Microsoft_Compiled_HTML_Help Weird. As if plain HTML with an ordinary browser would not be good enough for displaying a help text. That's M$ at its best (or worst). ------------------------------ Date: Fri, 13 Jun 2008 20:27:16 -0400 From: JF Mezei Subject: Re: What is a Weendoze .CHM file? Message-ID: <485311d0$0$7217$c3e8da3@news.astraweb.com> Michael Kraemer wrote: > Weird. As if plain HTML with an ordinary browser would not be > good enough for displaying a help text. > That's M$ at its best (or worst). Microsoft needs to find ways to justify bundling its virus magnet (internet explorer) with Windows and tell lawmakers that it is required. If that application is the only know that knows how to view this proprietary format, and all of the windows help is in that proprietary format, then guess which browser you are absolutely required to have to view the help files ? On the other hand, this isn't so different from Digital equipment corporation forcing you to use Bookreader to access their prorietary documentation format. Of course, this was a different era bad then, and decw$bookreader format was developped before HTML. ------------------------------ End of INFO-VAX 2008.330 ************************