ISS Debate

[dobbins]

The best way to get something accomplished with Congress is to ignore the Congress Critters and work with the baby sitters, the Congressional Aides. Half the Congress is nothing more than empty suits that the Aides carry around.<br /><br />

 

[radarredux]

> <i><font color="yellow">Assuming that Bigelow ever manages to produce something other than hot air.</font>/i><br /><br />Most bleeding-edge technology startups do fail. Sometimes the technology is too hard. Sometimes the customers don't show up (at least for your price point). Sometimes the foundational technologies aren't there.<br /><br />I read an article the other day saying that some Internet high tech companies are eschewing VC funding because the dramatically reduced hardware prices and mature open source software allows them to do things much cheaper today than they could in 2000. Sometimes market timing is important.</i>

 

[dobbins]

Don't get me wrong on one point, I hope that Bigelow and others succeed. I just think it's foolish to make plans on an assumption that they will succeed before they have demonstrated that they can do it.<br /><br />

 

[radarredux]

> <i><font color="yellow">ignore the Congress Critters and work with the baby sitters, the Congressional Aides.</font>/i><br /><br />Certainly any member cannot have indepth knowledge about all the range of topics they are expected to vote on, so they rely on their aides and lobbiests.</i>

 

[radarredux]

> <i><font color="yellow">I just think it's foolish to make plans on an assumption that they will succeed before they have demonstrated that they can do it.</font>/i><br /><br />That is pretty much what Griffin has said as well. The challenge will be: What will NASA do if the commercial sector does show up?</i>

 

[CalliArcale]

<blockquote><font class="small">In reply to:</font><hr /><p>Heck we are using Windows NT as our space operating system.<p><hr /></p></p></blockquote><br /><br />Actually, the dominant operating system in space is VxWorks, a POSIX-derivative produced as a COTS product by WindRiver. Various versions of Windows have been used in space (as well as Linux, I believe) to run non-critical laptop devices, but not much else. Older computers, suchas those in the Space Shuttle, use custom-designed operating systems.<br /><br />Believe me, I know. I work in space computing. <img src="/images/icons/wink.gif" /> My company's computers have flown on a lot of vehicles, including Chandra. We do not use Windows, because it does not remotely meet the requirements. Most significantly, it isn't even the right kind of an OS; it's not a real-time OS. <div class="Discussion_UserSignature"> <p> </p><p><font color="#666699"><em>"People assume that time is a strict progression of cause to effect, but actually from a non-linear, non-subjective viewpoint it's more like a big ball of wibbly wobbly . . . timey wimey . . . stuff."</em>  -- The Tenth Doctor, "Blink"</font></p> </div>

 

[erioladastra]

"Congress may have been focused on Katrina in September, but as an institution it has the attention span of a puppy."<br /><br />I don't disagree. I am just reitering what Griffin himself has said on his recent (October) meetings.

 

[frodo1008]

So Burt Rutan and Elon Musk are Russian supported???

 

[juliemac]

Just scrap the station and jump into a long trip in space with tools, machines and techniques that were never tested?<br /><br />We know that the Electron O2 generator has flaws. It does break down. Alpha test on Earth, Beta on the station. Abuse the heck out of the new designs till the systems and problems are well known. THEN use the system for a long trip.<br /><br />From playing in the shallows to a TransAtlantic crossing is just dangerous.<br /><br />Use the tools we have to make better and more reliable tools.

 

[erioladastra]

"Heck we are using Windows NT as our space operating system."<br /><br />Not in the case of ISS. The operating system is Ada for the station computers and Unix+XML (mostly) for the crew's interface laptop. The station LAN which is used for procedures, email etc is Windows.

 

[spacester]

OK I had a chance to read the thread from the top. Our story so far:<br /><br />I provided a thread for orrery to separate himself from unreasonable trolls and IMO he did a fine job of seizing the opportunity. I think I can sum up the central point in our personal interaction: We both agree that Bigelow is looking good to develop, over just a few years, a technology that in many ways will be superior to ISS. While I say that ISS will not be obsolete as a result, orrery maintains that it will be so badly obsolete as to not be worth doing.<br /><br />One of the reasons for starting this thread is that IMO we need to discuss the future role of ISS. The spin-gravity discussions were very helpful along these lines, at least to this spin-g proponent. <br /><br />But I still would like to be able to look squarely in the eye of the kill-the-ISS guys and tell them more exactly why they are wrong. I don’t think that can be done until a more detailed discussion takes place as to ISS capabilities and purposes.<br /><br />Here’s a well-written paragraph from orrery that I think needs more answers:<br /><br /><font color="yellow"> That the ISS will provide a plethora of capabilities is indeed a true observation. However, many and most of the Station capabilities will become redundant once a Bigelow station were to become operational. The question is whether the Station has value in being completed with just the few remaining unique capabilities. Bigelow modules will be lighter weight, and if NASA were to purchase a module it could easily launch it into a higher orbit... which would then give it a beneficial edge over the ISS. The ISS is stuck in low orbit... light weigh, cheap Bigelow modules could be purchased by the government and launched to almost any desired orbit which would present us with a trade-off scenario regarding the "unique capabilities" of the ISS and the Bigelow modules, and it is my belief that that debate will be won by those modules in higher orbit... (easier access to NE</font> <div class="Discussion_UserSignature"> </div>

 

[CalliArcale]

<blockquote><font class="small">In reply to:</font><hr /><p>Not in the case of ISS. The operating system is Ada for the station computers and Unix+XML (mostly) for the crew's interface laptop. The station LAN which is used for procedures, email etc is Windows.<p><hr /></p></p></blockquote><br /><br />*cringe*<br /><br />Ada and XML are not operating systems. Ada is a programming language; XML is a standard for markup of data in flat files. (HTML, in its current definition, is technically an implementation of XML, although some common but deprecated useage violates XML.) I believe the Shuttle computers don't have an OS at all; many embedded platforms do not require a general-purpose OS to run various programs, because they can instead be loaded with just the program itself. The program will be bigger than it would hav eto be if it could run on top of an OS, since it must itself include all of the functionality it would ordinarily expect from the OS, but the standalone program will be smaller than the smaller program + general purpose operating system, and it will run faster and more efficiently. Besides, these sorts of systems need to be real-time, and real-time OSes are relatively new (compared to the Shuttle's main computers, anyway -- Mars Pathfinder was the first spacecraft to run VxWorks).<br /><br />Ada is a popular programming language for space applications, thanks in large part to its widespread acceptance by the military. It is a very cumbersome language, but also very robust. It's much harder to write bad code in Ada than in C, because in many cases, bad code simply won't compile. <div class="Discussion_UserSignature"> <p> </p><p><font color="#666699"><em>"People assume that time is a strict progression of cause to effect, but actually from a non-linear, non-subjective viewpoint it's more like a big ball of wibbly wobbly . . . timey wimey . . . stuff."</em>  -- The Tenth Doctor, "Blink"</font></p> </div>

 

[cuddlyrocket]

"My suggestion is to talk about ISS Science and the future purposes of ISS so as to show that the ISS will have a place in our LEO future alongside Bigelow’s facilities."<br /><br />The ISS will have a place in our LEO future <i>because it is already there</i>. Bigelow's facilities <i>may</i> (it has yet to be demonstrated) be more cost-efficient when compared with building an ISS from scratch, but they are not more cost-effective when compared with utilising an existing facility.<br /><br />The ISS doesn't cost that much to maintain. It is likely to be true for quite a while that if you want to do something on a manned station, for a lot of purposes it will be cheaper to do it on the ISS.<br /><br />The main cost of the ISS in the future is completing the infrastructure, and that cost is the cost of the Shuttle flights - the hardware has already been paid for. No ISS construction no Shuttle (past one Hubble mission) - or to put it another way, if you're going to keep the Shuttle infrastructure going, you may as well use it to complete the ISS.<br /><br />Now, there's a perfectly respectable intellectual case to be made for scrapping both the ISS and the Shuttle now. Equally, there's a case for proceeding as NASA currently plans. But I just think that this is a pointless discussion. Political reality (which is just that, <i>reality</i>) is that Congress will not scrap the Shuttle immediately and will construct the ISS to justify not doing so.<br /><br />It is true that Congress will respond (however slowly and imperfectly) to the wishes of you-the-voter (and I mean 'you', as I'm English), but you-the-voter also includes all the people who will be sacked, and their families etc. But unlike those who are in favour of shutting down the Shuttle and scrapping the ISS, they have a direct stake and something to lose, and will therefore scream loudest. Perhaps more importantly, they are concentrated in a few districts where they can make a difference to the outcome.

 

[frodo1008]

Spacester: unlike a whole lot of the new and very negative threads on this forum, this thread makes an effort for informed debate, you are to be congratulated! <br /><br />As to your call for an informed position on the science to be done on the ISS you must realize that a very detailed analysis of this area can’t be had at this time. The ISS is still a long way from being totally complete. When it is completed (and I also agree completely with CuddlyRocket’s analysis, that the political reality is that it WILL be completed) it will have far more scientific experimental capability than any other such station will have for at least the next 10 to 20 years. There will be at least four complete scientific laboratories on board (another reason for the necessity of having at least six people on board at all times). The electrical generation capabilities will be vast in comparison to any other such station for a long time to come. <br /><br />Those who advocate using Bigelow’s modules instead should realize that Bigelow would not even have such modules if it weren’t for NASA’s original research. NASA researched the kind of modules that Bigelow is proposing as just a part of the ISS itself! If Bigelow can get a full sized module into a high Earth orbit, he has no present intention of using it for scientific research at all. He wishes to turn such modules into non gravity, non spin, types of small space Hotels for space tourism as this is the only relatively quick way to make a profit on the high costs of building and launching such modules in the first place! Unless NASA or other government space programs can come up with a whole lot of cash quickly Bigelow would have no choice but NOT to use his modules for scientific research. It IS simply economics. So the very purpose of the ISS and the Bigelow space modules (assuming Bigelow can even complete the modules and get them launched) is totally different from the scientific purpose of the ISS (regardless of what tha

 

[drwayne]

"It's much harder to write bad code in Ada than in C, because in many cases, bad code simply won't compile."<br /><br />I worked IV&V for 7 years on a program that proved it is very possible to take bad Fortran and turn it into bad Ada. <img src="/images/icons/frown.gif" /> <img src="/images/icons/wink.gif" /><br /><br />One thing for the community to keep in mind about computer hardware and software that is employed in space (and sometimes in defense as well)<br /><br />It will tend to be several years behind the commercial state of the art.<br /><br />This is due to the fact that such hardware (and software) must go through testing such as environmental testing that can take a year or two to perform. It must be able to withstand bit flips and bit locks that occur with exposure to radiation. Such testing takes time, and money, and specialized skill, which accounts in part for why what would appear to be "primitive" hardware appears on spacecraft, and costs as much as it does.<br /><br />That is why, in an age of Pentiums being well entrenched in the commercial market, I was seeing 286's commonly used.<br /><br />Wayne <div class="Discussion_UserSignature"> <p>"1) Give no quarter; 2) Take no prisoners; 3) Sink everything."  Admiral Jackie Fisher</p> </div>

 

[erioladastra]

The boot and executive software for the ISS computers (MDMs) is written in Ada. All operations are Ada programs except for Timeliner which is a pseudo program langauge developed by Draper but which is read by the executive running Ada. It is not the prettiest code but it gets the job done. The laptops were just recently converted to Unix which interfaces to the MDMs via mil-std-1553. Most (I think they are trying to make all but not sure if compelete yet) of the displays are XML. I think there is some C code in there too.

 

[erioladastra]

Only several years? <img src="/images/icons/smile.gif" /><br /><br />Yes, the ISS MDMs are 386s! Ugh. But they are in the process of being updated. Of course, by the time the Pentiums get up there we will be on Pentium 10s <img src="/images/icons/wink.gif" />

 

[spacester]

One thing I've wondered for years but haven't asked in a while: What if you just shield your processors better? Just shield the heck out of them.<br /><br />I mean even to the extreme of building large heavy thick walled enclosures that would normally be laughed at as space hardware. Might not that investment in mass be worth it in terms of capability and more importantly, reliability?<br /><br />There's going to be the occasional high energy particle (um, neutrinos?) that you can't do anything about, but we get those here on Earth anyway.<br /> <div class="Discussion_UserSignature"> </div>

 

[radarredux]

> <i><font color="yellow">What if you just shield your processors better? Just shield the heck out of them.</font>/i><br /><br />I made the mistake once of picking up a PC that could be used for sensitive military documents -- it was tempest tested. The shielding weighed a ton! It was in an unclassified area (no sensitive information was on it), so I assumed it was a regular PC. PCs in the 1980s were heavy enough, but that shielding was a shock for me.</i>

 

[erioladastra]

"I mean even to the extreme of building large heavy thick walled enclosures that would normally be laughed at as space hardware. Might not that investment in mass be worth it in terms of capability and more importantly, reliability? "<br /><br />That would only alleviate part of the problem. Pentiums, with their thinner chip making them more susceptible, have been cleared for space with no exrtra mass surrounding the computer. However, try writing code that is not vulernable is VERY hard and VERy expensive. You have to analyze every line of code and make sure that if a bit flips in there you won't suddenly open all your valves to space. And finally, now and at least for awhile, mass is money when it comes to launches.

 

[CalliArcale]

One solution (albeit one which does still require very rigorous software engineering) is processor voting. If one processor gets its data corrupted, that will be detected. Another thing you can do to help protect data travelling on a bus is to put in a lot of checkbits. I mean a LOT of checkbits. Your home PC may use one or two checkbits for a particular data word. Some spacecraft use more checkbits than there are bits in their data word! I think Galileo had a 32-bit bus, but still only had an 8-bit data word. This was because all the rest of the space was used up with checkbits -- three times as many checkbits as actual data-carrying bits! With a sufficiently rigorous error checking procedure, you can not only detect corrupted data but correct the corrupted bits. It's pretty cool. <div class="Discussion_UserSignature"> <p> </p><p><font color="#666699"><em>"People assume that time is a strict progression of cause to effect, but actually from a non-linear, non-subjective viewpoint it's more like a big ball of wibbly wobbly . . . timey wimey . . . stuff."</em>  -- The Tenth Doctor, "Blink"</font></p> </div>

 

[erioladastra]

"It's pretty cool."<br /><br />Correct. <img src="/images/icons/wink.gif" /> Code processing a little harder to do that with though so you have to test, test and test, which means $, $$ and $$$. But that is the cost of space exploration.

 

[mattblack]

THIS is the Station option (Option C) that I believe should have got built, in or out of a 50+degree inclination orbit.<br /><br />http://www.abo.fi/~mlindroo/Station/Slides/sld057.htm<br /> <div class="Discussion_UserSignature"> <p> </p><p>One Percent of Federal Funding For Space: America <strong><em><u>CAN</u></em></strong> Afford it!!  LEO is a <strong><em>Prison</em></strong> -- It's time for a <em><strong>JAILBREAK</strong></em>!!</p> </div>

 

[vt_hokie]

Ah, computer programming...one area where my skills are severely lacking, and something I find as enjoyable as a root canal. I guess that's one reason I want to move away from engineering...

 

[rubicondsrv]

double post deleted <div class="Discussion_UserSignature"> </div>