My bad .... I misread the QNX part. I thought they wanted to use QNX. Sorry. At one company I did a contract at, they use pSOS with a large code base ... so you are correct .. they won't change to QNX even though there is no new development in pSOS.<br>
<br>Vasili<br><br><div class="gmail_quote">On Sun, Aug 28, 2011 at 10:04 PM, Brandon Allbery <span dir="ltr"><<a href="mailto:allbery.b@gmail.com">allbery.b@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div dir="ltr"><div class="im">On Sun, Aug 28, 2011 at 22:33, Vasili I. Galchin <span dir="ltr"><<a href="mailto:vigalchin@gmail.com" target="_blank">vigalchin@gmail.com</a>></span> wrote:<br></div><div class="gmail_quote">
<div class="im"><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
1) The reason I said "over the top" is that QNX is highly optimized to bound kernel pathways. I was able to read kernel code. I have also worked on LynxOS and pSOS. Not dissing you <img goomoji="330" style="margin: 0pt 0.2ex; vertical-align: middle;" src="cid:330@goomoji.gmail"><br>
</blockquote><div><br></div></div><div>Sure, there are plenty of RTOSes out there. There were back then, too; there are still times when simpler environments are preferred (admittedly, some of them are more psychological than justifiable by the intended usage). I still think "over the top" is itself a bit over the top; we are far from being able to replace the immense number of traditional PLCs or embedded devices out there which nevertheless need to be retrofitted in software to support newer technologies. (See below; you have about 0% chance of getting someone to rip out all the existing PLCs on every segment of every track in the country and replace it with a computer running QNX or VXWorks. Even ignoring the [mostly labor, but with the number of devices in question even the hardware adds up] cost, consider the logistics; the segments that need it most are the ones you can't afford to take offline for any significant amount of time while refitting and testing the new gear.)</div>
<div><br></div><div>Now imagine that someone has a large deployed base of simple dedicated embedded hardware and needs to retrofit SMTP into it. That's a software-only change, and waving a manual while intoning "QNX" won't get you anywhere with the hardware.</div>
<div class="im">
<div> </div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> 2) What is the Haskell package that you are alluding to. I would like to know plus probably others on this list.<br>
</blockquote><div><br></div></div><div>Hm, I'm not sure it hit Hackage; I'm not seeing anything obvious there aside from vaguely recalling some changes to Atom to make it more suitable for code generation for PLCs. You might, however, take a look at <a href="http://cs.swan.ac.uk/%7Ecspj/docs/calco.pdf" target="_blank">http://cs.swan.ac.uk/~cspj/docs/calco.pdf</a> which uses Haskell SAT solvers to validate PLC ladder logic for track safety; I don't know how much of the rest of the system is in Haskell.</div>
<div><br></div></div><div class="im">-- <br>brandon s allbery <a href="mailto:allbery.b@gmail.com" target="_blank">allbery.b@gmail.com</a><br>wandering unix systems administrator (available) <a href="tel:%28412%29%20475-9364" value="+14124759364" target="_blank">(412) 475-9364</a> vm/sms<br>
<br>
</div></div>
</blockquote></div><br>