IMO, Puppy is vastly superior to Knoppix, and even many of the bigger-name distros. It's lean, fast, and Barry Kauler's built what may be one of the most technically excellent foundations of any distro out there. The newer versions of Puppy have the ability to use packages from other distros - the default at the moment is to use Ubuntu Lucid packages, but there are several other options available, and support for others is an ongoing thing.
Actually, for the question the OP posed, I would recommend Puppy over probably *any* other Linux distro...
Power over Ethernet is DEFINITELY the way this should go. For one thing, PoE and the PoE standard just works, and has now been tested, standardized, extended to higher power ranges (802.3af up to 12W, 802.3at to 25W (50W w/all wires), and is extremely interoperable across devices and manufacturers.
PoE is smart - The PSE (power supplying equipment) and PD (powered device) have to sync (mostly passively, it's clever) to determine power required and the presence of a PD before any power is delivered.
PoE also has the advantage of being the first non-trivial worldwide power interface (yep, I'm considering USB as trivial, for lots of reasons...), and its standardization on 48VDC means there are already zillions of telco and industrial devices (including all modern VoIP phones) that can easily interoperate with it, or be easily modified to use it.
If only the automotive guys hadn't had their heads up an locked before standardizing on 42VDC next-gen automotive electrical systems... (I can't wait for $500 batteries!)
Most of the recommendations here are for the hardcore do-it-yourselfers. I'm perfectly capable of writing good HTML/CSS/JavaScript code (Notepad++ or vi) - BUT, because I don't do it day in and day out, I'm slow and never wind up doing it nearly as well as I wish I had. It's really hard to stay up with this stuff if you dont' live and breathe it. (I run a kick-ass web app development team, but I'm not a programmer - even that's still not enough to really stay on top of this technology from a DIY perspective...)
I think that's why the OP is asking about power tools. I've just been looking at some myself, and Squarespace.com definitely stands out from the crowd. It's a very solid CMS/editor/web site builder/manager app with some very large and successful customers. Since they're a for-pay service (you can try it for free), and have many serious customers, they're unlikely to go away soon, like many free hosted solutions.
BTW, I have no relation with Squarespace other than as a very likely potential customer: I've just been looking for a tool that lets me easily host, build, and manage killer sites without having to keep up with all kinds of arcane technical crap, and Squarespace seems to be the best thing I've found to do exactly that. I'm planning on using them to build my next couple of sites, and may even do a couple more just because of the leverage I expect it'll give me...
Tcl is indeed an excellent choice for many embedded applications. It's not as trendy as many of the newer languages, but it is an awesome tool for getting a lot done with a very small amount of code (some joke that it stands for "Try Coding Less"), and with very minimal overhead.
Lua's certainly intriguing, and I might look at it for future projects, but the vast majority of what I've had my teams build in the past few years is Tcl (if small size is important), or Python (if it's less important).
Like Python, Tcl approaches executable pseudocode - it's almost always trivial to read a Tcl program and be able to follow it, unlike many other languages. The value of this for things that you (or others) have to maintain and/or extend over time is incalculable...
Tcl's biggest ding right now is that it doesn't have a "popular and modern" snazzy web app platform right now, although AOLserver has great integrated-from-the-ground-up Tcl support, and is definitely one of the apps that shows that Tcl can scale up as well as it scales down....
I need another laptop, but I'm not buying another one until I can avoid having to do all my work on a screen with the aspect ratio of a mail slot. I broke out my old ThinkPad 570E the other day, and its 13" screen seemed *huge* compared to the 13.3" HP dm3 I usually carry. Interestingly, both are only 768 px tall (not nearly enough - I'd really lve a portrait display laptop!), but the old IBM really felt like it had much more room, mostly because the screen is physically much taller, even if there aren't any more pixels there.
It's also gloriously non-reflective matte. If it were possible to stuff more RAM in it and get a modern network adapter for it, I'd seriously think about going back. (It's a sleeker package than many current laptops, and it has the best 2-stage media slice + desk base (handy if anyone ever needed 4(!) PCMCIA slots) docking station I've ever seen. (Unfortunately, only Windows is smart enough to be able to handle 2-stage docking correctly...)
Crappy widescreen aspect ratios are one big reason I'm thinking of a tablet - I'm just dying for vertical pixel count. I don't necessarily want a tablet, but there's simply no other ultraportable solution that can give me 1000+ pixels of vertical resolution at a sub-$700 price point...
Apple is definitely looking to strengthen their stranglehold on the OS X environment. This move makes it much harder to run OS X on non-Apple hardware - they'll make sure your system passes Apple genuine validation before you're allowed to download it.
Now I know that Apple's OS X license agreement says you can only run OS X on Apple hardware, but I also think that's an illegal restriction, and this move will make it nearly impossible to run O X on any hardware except what Apple has decided to allow you to run it on.
It'll be interesting to see what they do w.r.t. virtualization - will they allow VM images to only be downloaded to and run from Apple OS X Server instances? Do you now have to buy hideously expensive Apple server hardware to be able to benefit from virtualization?
This stinks to high heaven. I like Apple's products, but haven't been able to bring myself to buy them in several years - I just can't willingly march into the gulag...
The military's not "stuck", they're actually getting much slower. Compare the speeds of the top planes of the early 1960's to today's best military jets: Airplane Max. Speed, Mach B-58 Hustler 2.1 F-104 Starfighter 2.2 F-106A Delta Dart 2.5-2.85 B-70 Valkyrie 3.08 SR-71 Blackbird 3.3+ B-1 Lancer 1.25 F-14 Tomcat 2.3 F-15 Eagle 2.5+ F-16 Viper 2.0+ F-22 Raptor 1.7-1.8 (supercruise) F-35 Lightning II 1.6 (estimated)
As you can see, The speed of military aircraft has been on a general downward trend -current Air Force pilots can only dream of the speeds their grandfathers(!) flew
Trains are now easily twice the speed that they were a couple of decades ago (in places with decent rail systems) and they carry vastly more people than the shuttle or concorde. Even if you measure passenger-miles, these two are largely irrelevant. Making a subway train 50% faster has a much bigger impact on overall quality of life than making a transatlantic flight 50% faster. 5-10 minutes off a daily commute is a much bigger win than 2 hours off a 5 hour flight that most people are lucky to make once every few years.
Trains are irrelevant in the US, and always will be. The fundamental structure of geographic distribution of population is different in the US - no one, not even the railroad robber barons, ever made significant money on passenger rail in the US. Only freight can "carry the freight" to put a US railroad in the black. That's the reason why Amtrak had to be formed by the Feds: No railroad company wanted anything to do with passenger service. Amtrak is not only dog-slow, and run by uncaring union bureaucrats that make the post office look diligent, it also has an unblemished track record of losing huge quantities of taxpayer money year in and year out.
The best way to ensure the success of rail is to do just what the Europeans have done - it's as easy as 1,2,3: 1) Build rail at ridiculous public expense (after all, you've got the money, the US is subsidizing your defense!); 2) Jack the price of gasoline through the roof to ensure driving is too expensive, and; 3) Brutally control wages and regulate competition so the people can't afford other alternatives. Voila! Railroad success!
The cable companies are perhaps the greatest example today (with the possible exception of the Chevy Volt and the Nissan Leaf) of companies insisting on providing something other than what their customers really want. (Most Blockbuster customers didn't really give a rip about BluRay, but they let the studios talk them into billions in content "upgrades" while RedBox and Netflix kicked their butts with cheap DVDs...)
Unfortunately, the cablecos control most Internet access, too, so you can't really fire the bastards, even if you want to - and I *do* want to. I'm a TimeWarner customer here in Austin, and the service is truly awful. Since "upgrading" to digital cable service, I'm wishing for my old analog signal, which at least made it *possible* to watch a ballgame without everything dissolving into a pixellated mess with every pitch. (Seriously, analog cable really was *way* better.)
And don't get me started about TW's execrable DVR (which NOTHING like the real TiVo I gave up ) - I can honestly say this is perhaps the all-time worst performing and most thoughtlessly designed piece of um, consumer electronics I've ever seen. The UI is a train wreck (the defaults are frequently "dangerous" - for instance, the first/default option when recording a show is stop/delete! It takes seconds to respond when it decides to at all. It has no clue that it just recorded the same thing three times, so there's no intelligent way to have it do the "season pass" thing without actually picking each future showing by hand. I could go on, but it's BAD - really, really bad.
Worse, the city council protects TW through onerously expensive regulations for competitors - they're forced offer all high-end, high-speed services first in East Austin, where the take rates are so low that they can't afford to wire out the rest of the city. As a result, the choices are TW (awful), AT&T (even worse, with DSL dog-slow net), and Grande, which most can't get for the reasons above.
If I had a decent high-speed IP connection that didn't require TW, I'd drop cable in a heartbeat, except for one thing: There's no way yet to replace the lost cable channels, especially for news and opinion. If Fox News, CNN, CSPAN, Discovery, History Channel, et al were on Roku (essentially giving me the a la carte channel lineup the cableco refuses to), then the cable companies would be down ten million subscribers before a year was out. (Some have said sports is an issue, and while it's true that covering many sports is kinda pricey, I could get by fine with just MLB TV and rabbit ears for the occasional network broadcast I care about.)
When will we get a real ISP (NOT a cableco) with the vision to build out highspeed IPv6 net access with routers that support multicast streaming, so all those Rokus/Boxees/AppleTVs don't kill the net? The tech to do this has been available for a decade, and readily affordable for a couple of years, so what are we waiting for? The cable companies can't die quickly enough for me...
As someone who is hesitating to found a new manufacturing company here in the US, I can tell you the problem with mfg. here is NOT labor rates (at least not in right-to-work states where you can get better workers for far less than the rapacious union scales that have killed the auto, steel, and other US mfg industries) - the real problem, and the ONLY things keeping the US from competing globally in the manufacturing market, are: 1) the world's highest corporate tax rates, and 2) the outrageous and almost unpredictable cost of regulatory compliance. (Obamacare's 1099 reporting requirement is a new form of anti-small business evil, aimed at forcing every small business in America to move all transactions to credit/debit cards (for Obama's bailed-out bank cronies) or face literally crippling new paperwork, accounting, and reporting costs.
The US can be a manufacturing powerhouse again tomorrow - All we have to do is just roll back all the ridiculous regulations and the ever-increasing cost of trying to comply with the the myriad dictates of ever increasing armies of "bureaucrats armed and clerical" ("trying", because it's impossible to actually fully comply, by design.) We could start by completely eliminating several Federal departments or agencies that have NO backing or support in the US constitution. (Energy, Education, Commerce, Agriculture, OSHA, EPA, FDA, and Commerce would be a good start... That doesn't mean no one should do those functions, but that the Feds have NO business doing them - the 10th amendment makes that quite clear.)
If we eliminate punitive regulation and tax policies towards business, we'll see an entrepreneurial boom the like of which we haven't seen since the late industrial revolution - there's huge pent-up desire and will to build things here - we just need to get our government and its regulations the hell out of the way! (Has the number of pages of Federal regulations gone down *any* year in the last century? I highly doubt it, and Obama's "regulatory czar" Cass Sunstein is certainly working in the opposite direction.)
We should be willing to vote in a heartbeat for anyone who today would echo Barry Goldwater's famous words:
I have little interest in streamlining government or in making it more efficient, for I mean to reduce its size. I do not undertake to promote welfare, for I propose to extend freedom. My aim is not to pass laws, but to repeal them. It is not to inaugurate new programs, but to cancel old ones that do violence to the Constitution, or that have failed their purpose, or that impose on the people an unwarranted financial burden. I will not attempt to discover whether legislation is "needed" before I have first determined whether it is constitutionally permissible. And if I should later be attacked for neglecting my constituents' "interests," I shall reply that I was informed that their main interest is liberty and that in that cause I am doing the very best I can.
- The Conscience of A Conservative (1960)
While wget isn't capable of doing quite a lot of the "nice" work, curl definitely is capable of far more- I'm still amazed at how many people assume that wget is the best tool to use for grabbing web pages from the command line - I suppose it's a combination of ignorance and GNU bias.
Don't get me wrong, wget is not a bad program, but it's not even in the same league with curl. Curl is is not only far more capable, but also has a far more active developer community.
Curl is to web processors what ImageMagick is to image processors - the one indispensable command line tool you just don't want to be without, and a truly awesome tool to leverage in Taco Bell style development.
BTW, Taco Bell development is important for another reason as well - I am not a "real" programmer, but I can build real, solid apps very quickly using a set of well-chosen tools based on Unix text processing tools and the stream/operator paradigm. With modern hardware, it's surprising how well these "Taco Bell" apps can work - often well enough for production use in Fortune 500 environments.
Widescreens are the scourge of the early 21st century, and hopefully, will one day be looked on as being as ridiculous as the 5" monitors on early luggables or the nearly-as-useless 9" monitor that was Steve's idea of all we needed for half a decade.
Laptops make the problem worse - I recently bought a new laptop (hp dm3, an excellent machine except for an ill-advised effort to save cost by choosing a dodgy ALPS touchpad instead of Synaptics), but, like almost all laptops available now almost regardless of price, it's a step backwards in terms of VERTICAL pixel resolution. The dm3, like all other compact laptops on the market, sports a wimpy 1366x768 screen, and I really miss those extra 32 pixels from my previous 1280x800 screen. I will never again buy a screen less than 1K pixels high. At least give us the option of a 4:3 aspect ratio again. (I need a new desktop monitor, but I'm not buying one (or three) until I can get them in non-widescreen again!)
Vertical pixels are the only ones that actually have any value to most people: Most laptop users *never* watch movies on their PCs (that's why they have TVs, duh), and the activities people use laptops for almost always value screen height over width: web browsing, e-mail, document creation, PDF viewing, reading, and even spreadsheets, if they're used more for data organization than financial analysis, which I'd argue is the more common use.
I'd buy a portrait-format laptop in a heartbeat - The best thing about the iPad may well be that you can actually view an entire page on it at once. Apple realized that 1024 was the minimum pixel height for good page-viewing - and they also knew they'd be selling cases of them to companies just looking for a decent portable documentation viewer.
How about a laptop with a 14-15" portrait screen and the keyboard from a 12" netbook? I really don't think the notebook/netbook folks realize how much volume they stand to lose to the iPad and its ilk if they don't start producing screens people want to use.
Just rebuilding the Apollo "rocket to the moon" stuff would not be easy. Lots of the _unwritten_ knowledge has been lost - not everything is written down or can be. I won't be surprised if the records of what have been lost have been lost too:).
Tell me about it! True Story: I was just out of school and working my first job for a California aerospace company one morning in 1986 when the Challenger blew up. It wasn't two weeks before our entire group was in turmoil, because the Air Force had realized that the shuttle fleet was going to be grounded for years and they *had* to replace some critical spy sats in the not-too-distant future. Orders were issued to resurrect the Titan program and build some boosters to get by ASAP.
Enter "forensic engineering" - we had the drawings in storage, but they had been chewed up pretty severely by mice - we were literally piecing them together with tweezers and tape in a scene out of NCIS, and guessing at what was on the areas that were flat missing.
In one memorable instance (the tool for the carbon-carbon exhaust nozzle) we had tooling, but had no idea how to use it - the plies laid in on a spiral, so it was easy to get started, but we had no idea how we were supposed to wedge the last several dozen plies in place. The shop-floor documentation said what to do, but gave no clue as to how to do it! Whoever built it knew things we didn't - but most of the people who had worked on the program were retired, dead, or addled, so we were pretty much on our own. I left before the company figured it out - it's quite possible they had to build a new multipart tool - either that or they finally found an old-timer to clue them in, since the Titan flew on schedule not too much later...
If I'm understanding the original poster correctly, one of the things he wants to do is have mulitple viewers of a single transmitter stream (possibly with some kind of moderated talk-back capability, too).
I'm far from an expert in this area (I haven't worked on a videoconferencing system selection in nearly a decade, and don't really use video chat myself - heck, I don't even *have* a skype account), but it seems like the OP needs some kind of point-to-multipoint capability.
Do any of the suggested solutions actually support that in any useful way?
You simply can't get all of what you want, but here's my advice, based on building a number of mobile networks for a variety of business clients:
Forget satellite unless you *really* need connectivity anywhere. It's expensive, complicated, expensive, the latency sucks, and it's expensive. Seriously, if you want to remain connected in Southeast Louisiana (which we judged as "more remote" than the Arabian desert for one project), then satellite's your solution, but the drawbacks are such that even then, it's an adjunct, not a primary solution - if any other network is available, you'll want to use that instead, anyway. Oh, and did I mention it's really, really expensive?
Wireless 3G is your best bet right now. Based on real-world experience (designing networks for deployment to restaurants nationwide), I *strongly* suggest a CDMA network: Sprint first, followed by Verizon as #2. I looked and tried a number of options, and Sprint definitely offered the best chance of a stable, high-speed network connection. Verizon wasn't bad, but fast connections seem much harder to come by on Verizon than Sprint, especially in remote areas. (Also, remember that Sprint now offers combined 3G/4G aircards, so you can use WiMax, if you ever expect to find yourself in a WiMax city. I'd be thrilled just to have Austin as a WiMax city, myself...) Avoid AT&T like the plague of iPhones that has their network gasping for its last breath. Seriously, even before the iPhone, AT&T's data network was far worse than Sprint's, and although it's much better now, I think they've slipped even further due to not keeping up with the iPhone hordes. IMHO, T-mobile and other carriers don't really have the broad footprint required to be an "almost anywhere" solution. You'll still have no connection in remote areas, but the only fix for that is satellite, with all the problems noted above.
I've deployed many Cradlepoint MBR-1000s, and recommend them highly. This is the "professional" approach as opposed to the "consumer" type MiFi devices - this approach gives you much more smarts in the router, plus using an external aircard lets you add an external 3G antenna, which can make a *huge* difference. Yes, you could also roll your own and save a little bit. No, it's not worth it. IMO, the Cradlepoint is worth twice what it costs.
With a good external 3G antenna, the MBR-1000 and a good aircard is the best solution available today for remote high-speed Internet. BTW, I recommend Sierra Wireless aircards for a couple of reasons: 1) They work better with the Cradlepoints than other brands, and 2) some Novatel aircards are a royal PITA to set up. I've had good experiences with 3Gstore.com for external antennas and Cradlepoint support. Make sure your aircard supports external antenna connections, and that you can get a connector for it (these are far from standard right now). Finally, although 3G latencies are orders of magnitude better than satellite, they are still much higher than landline networks - VoIP isn't really practical, for instance. Fortunately, these networks natively support voice traffic for cellphones as well, so it's really not much of an issue.;-) BTW, speed on these networks is pretty decent - in my last startup, we used a Sprint 3G/Cradlepoint connection as our primary Internet connection for a small office (3-6 users) for a year and a half, with only occasional bandwidth snarls.
I'd never want to travel again without my Cradlepoint PHS (kind of like the MiFi, but uses an external USB aircard)- it's just way too handy to be doing the multitasking-in-the-car thing like what's shown in that new Sprint commercial...
I'm pretty sure the older version of 568A (the standard itself, NOT to be confused with the T568A or T568B termination pair assignments) did mention this.
As mentioned in my post above, it's not optimal, but if you've already got wire in the walls and it's really hard/expensive to run new cable, this used to make sense. (It makes a lot less sense now with today's higher speeds, but is still applicable to many homes like mine, where running new wires involves lots of hole-cutting and wall and ceiling patching/painting, making it prohibitively expensive.)
I've never seen network electronics designed to work with such a setup, and as mentioned above, it really makes termination a bear, since you have to strip out the voice pairs and take them to a different set of punchdown blocks.
Although you *can* plug an RJ-11 into an RJ-45, every time I've ever seen this done, the outlets are terminated with two jacks: one RJ-45 (four-wires for Ethernet), and one RJ-11 for the middle voice pair.
It all works just fine, but there are good reasons you don't see it as a normative practice.
the idea of running a 48VDC pair that also uses a 100VAC ring signal right beside your ethernet pairs is scary
The 48V *DC* loop induces no noise at all, and in fact, since the positive side is at ground potential, it may actually help a tiny bit. 802.3af Power-over-Ethernet uses this exact same voltage, but puts it over either the "spare" pairs (there aren't any in GigE), or on the data pairs themselves - all PoE PD (powered device) implementations must support both. BTW, it's Ground and -48VDC so that the telco can put a sacrificial anode ground rod somewhere to eliminate corrosion problems - this isn't new, it's been that way for well over a century.)
Finally, the telco ring signal is a nominal 90VDC square wave, and I've designed wiring architectures for entire building campuses that used phone and Ethernet on pairs in the same jacket to avoid the huge cost of running new cable. I can honestly say I don't think I've ever seen the ring signal cause any significant data problems, unless there's something else wrong, too. Sure, it's best to separate voice and data if you can (mainly because of termination issues), but they work just fine together. (The worst problem is that the sides of the RJ-11 can sometimes spring the contacts for pins 1/2 and 7/8.)
Twisted Pair Ethernet is *incredibly* resilient - you actually *can* run it over rusty barbed wire - someone (SynOptics, maybe?) showed a cross-booth rusty barbed wire TP Ethernet link at Interop back in '93 or '94, IIRC. ('Course, that was back in the 10 Mbps days, with the old Manchester encoding, so it's about bulletproof...) Really, it's the same with any digital signal - I always laugh at people buying "Monster"-style HDMI cables - hell, it's a square wave: it either gets there or it doesn't, and expensive cabling will only make a difference if something is horribly wrong somewhere else!
In any case, if you use quality punchdown blocks and are careful not to blow the length or the dB budget for connections and terminations, it's hard to build an TP Enet that *won't * work, whether there are voice pairs in the same jacket or not...
P.S.: You're not really all that likely to feel the 48V unless you lick the wires, although it is enough to spark. Holding the wires when a ring signal comes through will get your attention, but that's true for any phone wiring...
Word is a word processor, not a text processor. There's a big difference: Word processors (Word, OOWriter, WordPerfect, AbiWord, etc.) mix up content and formatting, much like in the old pre-CSS HTML days. Formatting in WPs is generally specific text markup (bold, size, etc.) applied to a specific range. Text processors (most commonly the *TeX and *roff families) separate most formatting from the content - you just pour your text into the various content types and the text processor will apply whatever format it's been told. TPs tend to be more tag, style, or attribute-based - WPs with styles are somewhere in between, but as many have noted, actually using and enforcing correct use of styles or other predefined formatting in most WPs is difficult to impossible, especially with multiple editors/authors. TP documents are harder to get set up (TeX in particular is a PITA), but once that's done, it's much easier to focus on content and forget formatting.
Final thoughts: 1. Word isn't going away unless GoogleDocs or the like kills it, since it's still the only rich text editing tool of any consequence that is widely available, and it has the advantage of being sort-of compatible with itself. 2. Putting so much critical and nearly uneditable information in that little (normally) invisible paragraph mark at the end of your paragraph is simply insane!
Nobody's nailed the *real* reason most people choose PDF for document distribution: It's still the only foolproof way (for most people) to make sure that WIS (what **I** see) is WYG (what **YOU** get). If fonts and layout matter, then PDF lives simply because it can nail those down.
Word docs are awful for transferring documents, since if the recipient has another version of Word, or different/missing font, then all bets are off as to what will be at the bottom of page 8. (OpenOffice takes hits from both directions here, since it is both a varying version and its fonts are different.)
I *hate* the idea of PDFs, but I use them extensively, because the font industry's insistence on DRM to prevent font proliferation has screwed up documents for the last 20 years. Hell, we're only just now starting to *maybe* be able to use fonts and typography on the web for the first time, but it's far from certain that the foundries won't take steps to shut this down.
And what, exactly is really the difference between embedding a font in a PDF and embedding a font in a word processor document file? The former is legal, the latter is not. I guess it's that unbreakable Adobe Acrobat encryption or something...;-)
There's no fundamental reason MS Word can't be a great program. All it needs is a pioneering visionary to thrash it down to a working core, to develop some well thought out guiding principles for how to organize the interface, to mercilessly eliminate the rampant bugs, to study how the current interface fails, and to rebuild it from that working core back up to a well-engineered product. But will that happen? Unlikely.
But in reality, as bad as Word's faults are, Word has seen far more of this sort of attention (especially with the new "ribbon" UI in Word2007) than any of its competitors. Although I think Word's UI is still light-years short of where it should be, it's also light-years *ahead* of it competitors, including OpenOffice, which from a UI perspective is a bad copy of Word's interface circa Word 4.3/95 with a dash of Word2000.
In fact, although I don't agree with many of the decisions MS made for the new Office ribbon interfaces, there's no question that they *really* did the work to completely re-think the interface. They even posted pretty detailed explanations on the web (Chan9, IIRC) describing how they reached the major decisions, and why the choice they made seemed to strike the best balance.
Also, keep in mind that MS is severely hampered in what it can do by its user base, who are extremely sensitive to change of any kind (inertia, training costs, institutional stupidity, macros, etc.), even if that change results in a far better product.
And Microsoft is at least *trying* to think about user experience and common look and feel across its applications, something that OO seems to be relatively incapable of - screaming about the UI dates back to the StarDivision days. It's not that OO doesn't suck, it's just that it sucks less than any other free alternative to Word. (WordPerfect also has its flaws, but it absolutely kicks Word's butt for stability in producing really large, complex documents, which is why it's been the legal industry's favorite WP until very recently.)
Give the money in well structured grants to the private sector, like Burt Ruttan and Elon Musk, at least they are smaller, leaner and willing to think outside the box
Actually, those two guys may be the biggest reason NASA wants to bring down the entirety of a station it only barely owns half of in the first place!
Think this out a bit:
1. Once the shuttle retires, NASA will have no manned spaceflight capabilities to speak of, which the ISS requires to stay up. 2. Because of 1, NASA would have to "abandon" the ISS by leaving it unmanned. 3. Entrepreneurs and inventors really love a challenge, and a prize. And the ISS is quite a prize. 4. Getting someone aboard the ISS may well be legally "taking possession" of it. (I'm making the plausible assumption that the salvage laws in space would be found to be the same as, or largely similar to, those of the seas.) 5. The ISS cabal definitely doesn't want a spaceborne Sealand, and they'd rather torch a half trillion taxpayer bucks than let that even be a possibility. Things get worse if you contemplate unfriendly countries occupying the ISS, possibly as an orbiting recon and weapons platform. (NORK-ISS, anyone?;-))
This may be a far more credible rationale for "de-orbit" than any bogus "safety" argument: NASA and the other ISS owners can't keep occupying it, which is the only thing that perpetuates "ownership", and they can't stand the thought of anyone else getting to own it, so they'd rather destroy it.
But it's interesting that MS is making Windows 7 hard to use on netbooks and older laptops don't you think? That's because of something I haven't even seen anyone mention - the Win 7 RC is *only* available on DVD. This has been enough to deter me from trying it, since every single machine I would want to try it on does not have a DVD drive. This also says something about the bloat of Win7 as opposed to XP - I'd really like for the MS folks to take a serious look at Puppy Linux to see how you can build a small, light, fast, highly functional OS that fits in just a couple of hundred MB, even with a rich app set. (XP Puppy - yeah!)
Heck, for that matter, I don't even own a DVD burner, so I couldn't write the disc here at home even if the PCs and laptops I'd like to try it on could read a DVD. This is simply because movies are the absolute last thing I ever want to do on a computer, and I stopped using optical for backups back when the size of a CD was adequate for most things.
Good point. I agree Tufte has many acolytes who simply parrot whatever he says, and that much of what is says actually *is* simply a pronouncement of his opinion, often without sufficient analytical thought to back it up.
That said, Tufte's a good writer, a decent thinker, and has been right often enough and on enough topics to have built a certain degree of credibility. Personally, I think "The Visual Display of Quantitative Information" is indeed a great nail on which to hang a career, but his ideas on user interfaces and the like are considerably less well-thought-out (especially sparklines, which are useful in some cases, but certainly not the cure-all he seems to think).
FWIW, I have never taken any of his courses, and don't even own any of his books, although Visual Display is on my wanted list.
Slashdot Mirror
IMO, Puppy is vastly superior to Knoppix, and even many of the bigger-name distros. It's lean, fast, and Barry Kauler's built what may be one of the most technically excellent foundations of any distro out there. The newer versions of Puppy have the ability to use packages from other distros - the default at the moment is to use Ubuntu Lucid packages, but there are several other options available, and support for others is an ongoing thing.
Actually, for the question the OP posed, I would recommend Puppy over probably *any* other Linux distro...
Power over Ethernet is DEFINITELY the way this should go. For one thing, PoE and the PoE standard just works, and has now been tested, standardized, extended to higher power ranges (802.3af up to 12W, 802.3at to 25W (50W w/all wires), and is extremely interoperable across devices and manufacturers.
PoE is smart - The PSE (power supplying equipment) and PD (powered device) have to sync (mostly passively, it's clever) to determine power required and the presence of a PD before any power is delivered.
PoE also has the advantage of being the first non-trivial worldwide power interface (yep, I'm considering USB as trivial, for lots of reasons...), and its standardization on 48VDC means there are already zillions of telco and industrial devices (including all modern VoIP phones) that can easily interoperate with it, or be easily modified to use it.
If only the automotive guys hadn't had their heads up an locked before standardizing on 42VDC next-gen automotive electrical systems... (I can't wait for $500 batteries!)
Most of the recommendations here are for the hardcore do-it-yourselfers. I'm perfectly capable of writing good HTML/CSS/JavaScript code (Notepad++ or vi) - BUT, because I don't do it day in and day out, I'm slow and never wind up doing it nearly as well as I wish I had. It's really hard to stay up with this stuff if you dont' live and breathe it. (I run a kick-ass web app development team, but I'm not a programmer - even that's still not enough to really stay on top of this technology from a DIY perspective...)
I think that's why the OP is asking about power tools. I've just been looking at some myself, and Squarespace.com definitely stands out from the crowd. It's a very solid CMS/editor/web site builder/manager app with some very large and successful customers. Since they're a for-pay service (you can try it for free), and have many serious customers, they're unlikely to go away soon, like many free hosted solutions.
BTW, I have no relation with Squarespace other than as a very likely potential customer: I've just been looking for a tool that lets me easily host, build, and manage killer sites without having to keep up with all kinds of arcane technical crap, and Squarespace seems to be the best thing I've found to do exactly that. I'm planning on using them to build my next couple of sites, and may even do a couple more just because of the leverage I expect it'll give me...
Tcl is indeed an excellent choice for many embedded applications. It's not as trendy as many of the newer languages, but it is an awesome tool for getting a lot done with a very small amount of code (some joke that it stands for "Try Coding Less"), and with very minimal overhead.
Lua's certainly intriguing, and I might look at it for future projects, but the vast majority of what I've had my teams build in the past few years is Tcl (if small size is important), or Python (if it's less important).
Like Python, Tcl approaches executable pseudocode - it's almost always trivial to read a Tcl program and be able to follow it, unlike many other languages. The value of this for things that you (or others) have to maintain and/or extend over time is incalculable...
Tcl's biggest ding right now is that it doesn't have a "popular and modern" snazzy web app platform right now, although AOLserver has great integrated-from-the-ground-up Tcl support, and is definitely one of the apps that shows that Tcl can scale up as well as it scales down....
I need another laptop, but I'm not buying another one until I can avoid having to do all my work on a screen with the aspect ratio of a mail slot. I broke out my old ThinkPad 570E the other day, and its 13" screen seemed *huge* compared to the 13.3" HP dm3 I usually carry. Interestingly, both are only 768 px tall (not nearly enough - I'd really lve a portrait display laptop!), but the old IBM really felt like it had much more room, mostly because the screen is physically much taller, even if there aren't any more pixels there.
It's also gloriously non-reflective matte. If it were possible to stuff more RAM in it and get a modern network adapter for it, I'd seriously think about going back. (It's a sleeker package than many current laptops, and it has the best 2-stage media slice + desk base (handy if anyone ever needed 4(!) PCMCIA slots) docking station I've ever seen. (Unfortunately, only Windows is smart enough to be able to handle 2-stage docking correctly...)
Crappy widescreen aspect ratios are one big reason I'm thinking of a tablet - I'm just dying for vertical pixel count. I don't necessarily want a tablet, but there's simply no other ultraportable solution that can give me 1000+ pixels of vertical resolution at a sub-$700 price point...
Apple is definitely looking to strengthen their stranglehold on the OS X environment. This move makes it much harder to run OS X on non-Apple hardware - they'll make sure your system passes Apple genuine validation before you're allowed to download it.
Now I know that Apple's OS X license agreement says you can only run OS X on Apple hardware, but I also think that's an illegal restriction, and this move will make it nearly impossible to run O X on any hardware except what Apple has decided to allow you to run it on.
It'll be interesting to see what they do w.r.t. virtualization - will they allow VM images to only be downloaded to and run from Apple OS X Server instances? Do you now have to buy hideously expensive Apple server hardware to be able to benefit from virtualization?
This stinks to high heaven. I like Apple's products, but haven't been able to bring myself to buy them in several years - I just can't willingly march into the gulag...
The military's not "stuck", they're actually getting much slower. Compare the speeds of the top planes of the early 1960's to today's best military jets:
Airplane Max. Speed, Mach
B-58 Hustler 2.1
F-104 Starfighter 2.2
F-106A Delta Dart 2.5-2.85
B-70 Valkyrie 3.08
SR-71 Blackbird 3.3+
B-1 Lancer 1.25
F-14 Tomcat 2.3
F-15 Eagle 2.5+
F-16 Viper 2.0+
F-22 Raptor 1.7-1.8 (supercruise)
F-35 Lightning II 1.6 (estimated)
As you can see, The speed of military aircraft has been on a general downward trend -current Air Force pilots can only dream of the speeds their grandfathers(!) flew
Trains are now easily twice the speed that they were a couple of decades ago (in places with decent rail systems) and they carry vastly more people than the shuttle or concorde. Even if you measure passenger-miles, these two are largely irrelevant. Making a subway train 50% faster has a much bigger impact on overall quality of life than making a transatlantic flight 50% faster. 5-10 minutes off a daily commute is a much bigger win than 2 hours off a 5 hour flight that most people are lucky to make once every few years.
Trains are irrelevant in the US, and always will be. The fundamental structure of geographic distribution of population is different in the US - no one, not even the railroad robber barons, ever made significant money on passenger rail in the US. Only freight can "carry the freight" to put a US railroad in the black. That's the reason why Amtrak had to be formed by the Feds: No railroad company wanted anything to do with passenger service. Amtrak is not only dog-slow, and run by uncaring union bureaucrats that make the post office look diligent, it also has an unblemished track record of losing huge quantities of taxpayer money year in and year out.
The best way to ensure the success of rail is to do just what the Europeans have done - it's as easy as 1,2,3: 1) Build rail at ridiculous public expense (after all, you've got the money, the US is subsidizing your defense!); 2) Jack the price of gasoline through the roof to ensure driving is too expensive, and; 3) Brutally control wages and regulate competition so the people can't afford other alternatives. Voila! Railroad success!
The cable companies are perhaps the greatest example today (with the possible exception of the Chevy Volt and the Nissan Leaf) of companies insisting on providing something other than what their customers really want. (Most Blockbuster customers didn't really give a rip about BluRay, but they let the studios talk them into billions in content "upgrades" while RedBox and Netflix kicked their butts with cheap DVDs...)
Unfortunately, the cablecos control most Internet access, too, so you can't really fire the bastards, even if you want to - and I *do* want to. I'm a TimeWarner customer here in Austin, and the service is truly awful. Since "upgrading" to digital cable service, I'm wishing for my old analog signal, which at least made it *possible* to watch a ballgame without everything dissolving into a pixellated mess with every pitch. (Seriously, analog cable really was *way* better.)
And don't get me started about TW's execrable DVR (which NOTHING like the real TiVo I gave up ) - I can honestly say this is perhaps the all-time worst performing and most thoughtlessly designed piece of um, consumer electronics I've ever seen. The UI is a train wreck (the defaults are frequently "dangerous" - for instance, the first/default option when recording a show is stop/delete! It takes seconds to respond when it decides to at all. It has no clue that it just recorded the same thing three times, so there's no intelligent way to have it do the "season pass" thing without actually picking each future showing by hand. I could go on, but it's BAD - really, really bad.
Worse, the city council protects TW through onerously expensive regulations for competitors - they're forced offer all high-end, high-speed services first in East Austin, where the take rates are so low that they can't afford to wire out the rest of the city. As a result, the choices are TW (awful), AT&T (even worse, with DSL dog-slow net), and Grande, which most can't get for the reasons above.
If I had a decent high-speed IP connection that didn't require TW, I'd drop cable in a heartbeat, except for one thing: There's no way yet to replace the lost cable channels, especially for news and opinion. If Fox News, CNN, CSPAN, Discovery, History Channel, et al were on Roku (essentially giving me the a la carte channel lineup the cableco refuses to), then the cable companies would be down ten million subscribers before a year was out. (Some have said sports is an issue, and while it's true that covering many sports is kinda pricey, I could get by fine with just MLB TV and rabbit ears for the occasional network broadcast I care about.)
When will we get a real ISP (NOT a cableco) with the vision to build out highspeed IPv6 net access with routers that support multicast streaming, so all those Rokus/Boxees/AppleTVs don't kill the net? The tech to do this has been available for a decade, and readily affordable for a couple of years, so what are we waiting for? The cable companies can't die quickly enough for me...
As someone who is hesitating to found a new manufacturing company here in the US, I can tell you the problem with mfg. here is NOT labor rates (at least not in right-to-work states where you can get better workers for far less than the rapacious union scales that have killed the auto, steel, and other US mfg industries) - the real problem, and the ONLY things keeping the US from competing globally in the manufacturing market, are: 1) the world's highest corporate tax rates, and 2) the outrageous and almost unpredictable cost of regulatory compliance. (Obamacare's 1099 reporting requirement is a new form of anti-small business evil, aimed at forcing every small business in America to move all transactions to credit/debit cards (for Obama's bailed-out bank cronies) or face literally crippling new paperwork, accounting, and reporting costs.
The US can be a manufacturing powerhouse again tomorrow - All we have to do is just roll back all the ridiculous regulations and the ever-increasing cost of trying to comply with the the myriad dictates of ever increasing armies of "bureaucrats armed and clerical" ("trying", because it's impossible to actually fully comply, by design.) We could start by completely eliminating several Federal departments or agencies that have NO backing or support in the US constitution. (Energy, Education, Commerce, Agriculture, OSHA, EPA, FDA, and Commerce would be a good start... That doesn't mean no one should do those functions, but that the Feds have NO business doing them - the 10th amendment makes that quite clear.)
If we eliminate punitive regulation and tax policies towards business, we'll see an entrepreneurial boom the like of which we haven't seen since the late industrial revolution - there's huge pent-up desire and will to build things here - we just need to get our government and its regulations the hell out of the way! (Has the number of pages of Federal regulations gone down *any* year in the last century? I highly doubt it, and Obama's "regulatory czar" Cass Sunstein is certainly working in the opposite direction.)
We should be willing to vote in a heartbeat for anyone who today would echo Barry Goldwater's famous words:
While wget isn't capable of doing quite a lot of the "nice" work, curl definitely is capable of far more- I'm still amazed at how many people assume that wget is the best tool to use for grabbing web pages from the command line - I suppose it's a combination of ignorance and GNU bias.
Don't get me wrong, wget is not a bad program, but it's not even in the same league with curl. Curl is is not only far more capable, but also has a far more active developer community.
Curl is to web processors what ImageMagick is to image processors - the one indispensable command line tool you just don't want to be without, and a truly awesome tool to leverage in Taco Bell style development.
BTW, Taco Bell development is important for another reason as well - I am not a "real" programmer, but I can build real, solid apps very quickly using a set of well-chosen tools based on Unix text processing tools and the stream/operator paradigm. With modern hardware, it's surprising how well these "Taco Bell" apps can work - often well enough for production use in Fortune 500 environments.
Widescreens are the scourge of the early 21st century, and hopefully, will one day be looked on as being as ridiculous as the 5" monitors on early luggables or the nearly-as-useless 9" monitor that was Steve's idea of all we needed for half a decade.
Laptops make the problem worse - I recently bought a new laptop (hp dm3, an excellent machine except for an ill-advised effort to save cost by choosing a dodgy ALPS touchpad instead of Synaptics), but, like almost all laptops available now almost regardless of price, it's a step backwards in terms of VERTICAL pixel resolution. The dm3, like all other compact laptops on the market, sports a wimpy 1366x768 screen, and I really miss those extra 32 pixels from my previous 1280x800 screen. I will never again buy a screen less than 1K pixels high. At least give us the option of a 4:3 aspect ratio again. (I need a new desktop monitor, but I'm not buying one (or three) until I can get them in non-widescreen again!)
Vertical pixels are the only ones that actually have any value to most people: Most laptop users *never* watch movies on their PCs (that's why they have TVs, duh), and the activities people use laptops for almost always value screen height over width: web browsing, e-mail, document creation, PDF viewing, reading, and even spreadsheets, if they're used more for data organization than financial analysis, which I'd argue is the more common use.
I'd buy a portrait-format laptop in a heartbeat - The best thing about the iPad may well be that you can actually view an entire page on it at once. Apple realized that 1024 was the minimum pixel height for good page-viewing - and they also knew they'd be selling cases of them to companies just looking for a decent portable documentation viewer.
How about a laptop with a 14-15" portrait screen and the keyboard from a 12" netbook? I really don't think the notebook/netbook folks realize how much volume they stand to lose to the iPad and its ilk if they don't start producing screens people want to use.
DEATH TO WIDESCREENS!
Just rebuilding the Apollo "rocket to the moon" stuff would not be easy. Lots of the _unwritten_ knowledge has been lost - not everything is written down or can be. I won't be surprised if the records of what have been lost have been lost too :).
Tell me about it! True Story: I was just out of school and working my first job for a California aerospace company one morning in 1986 when the Challenger blew up. It wasn't two weeks before our entire group was in turmoil, because the Air Force had realized that the shuttle fleet was going to be grounded for years and they *had* to replace some critical spy sats in the not-too-distant future. Orders were issued to resurrect the Titan program and build some boosters to get by ASAP.
Enter "forensic engineering" - we had the drawings in storage, but they had been chewed up pretty severely by mice - we were literally piecing them together with tweezers and tape in a scene out of NCIS, and guessing at what was on the areas that were flat missing.
In one memorable instance (the tool for the carbon-carbon exhaust nozzle) we had tooling, but had no idea how to use it - the plies laid in on a spiral, so it was easy to get started, but we had no idea how we were supposed to wedge the last several dozen plies in place. The shop-floor documentation said what to do, but gave no clue as to how to do it! Whoever built it knew things we didn't - but most of the people who had worked on the program were retired, dead, or addled, so we were pretty much on our own. I left before the company figured it out - it's quite possible they had to build a new multipart tool - either that or they finally found an old-timer to clue them in, since the Titan flew on schedule not too much later...
If I'm understanding the original poster correctly, one of the things he wants to do is have mulitple viewers of a single transmitter stream (possibly with some kind of moderated talk-back capability, too).
I'm far from an expert in this area (I haven't worked on a videoconferencing system selection in nearly a decade, and don't really use video chat myself - heck, I don't even *have* a skype account), but it seems like the OP needs some kind of point-to-multipoint capability.
Do any of the suggested solutions actually support that in any useful way?
You simply can't get all of what you want, but here's my advice, based on building a number of mobile networks for a variety of business clients:
Forget satellite unless you *really* need connectivity anywhere. It's expensive, complicated, expensive, the latency sucks, and it's expensive. Seriously, if you want to remain connected in Southeast Louisiana (which we judged as "more remote" than the Arabian desert for one project), then satellite's your solution, but the drawbacks are such that even then, it's an adjunct, not a primary solution - if any other network is available, you'll want to use that instead, anyway. Oh, and did I mention it's really, really expensive?
Wireless 3G is your best bet right now. Based on real-world experience (designing networks for deployment to restaurants nationwide), I *strongly* suggest a CDMA network: Sprint first, followed by Verizon as #2. I looked and tried a number of options, and Sprint definitely offered the best chance of a stable, high-speed network connection. Verizon wasn't bad, but fast connections seem much harder to come by on Verizon than Sprint, especially in remote areas. (Also, remember that Sprint now offers combined 3G/4G aircards, so you can use WiMax, if you ever expect to find yourself in a WiMax city. I'd be thrilled just to have Austin as a WiMax city, myself...) Avoid AT&T like the plague of iPhones that has their network gasping for its last breath. Seriously, even before the iPhone, AT&T's data network was far worse than Sprint's, and although it's much better now, I think they've slipped even further due to not keeping up with the iPhone hordes. IMHO, T-mobile and other carriers don't really have the broad footprint required to be an "almost anywhere" solution. You'll still have no connection in remote areas, but the only fix for that is satellite, with all the problems noted above.
I've deployed many Cradlepoint MBR-1000s, and recommend them highly. This is the "professional" approach as opposed to the "consumer" type MiFi devices - this approach gives you much more smarts in the router, plus using an external aircard lets you add an external 3G antenna, which can make a *huge* difference. Yes, you could also roll your own and save a little bit. No, it's not worth it. IMO, the Cradlepoint is worth twice what it costs.
With a good external 3G antenna, the MBR-1000 and a good aircard is the best solution available today for remote high-speed Internet. BTW, I recommend Sierra Wireless aircards for a couple of reasons: 1) They work better with the Cradlepoints than other brands, and 2) some Novatel aircards are a royal PITA to set up. I've had good experiences with 3Gstore.com for external antennas and Cradlepoint support. Make sure your aircard supports external antenna connections, and that you can get a connector for it (these are far from standard right now). Finally, although 3G latencies are orders of magnitude better than satellite, they are still much higher than landline networks - VoIP isn't really practical, for instance. Fortunately, these networks natively support voice traffic for cellphones as well, so it's really not much of an issue. ;-) BTW, speed on these networks is pretty decent - in my last startup, we used a Sprint 3G/Cradlepoint connection as our primary Internet connection for a small office (3-6 users) for a year and a half, with only occasional bandwidth snarls.
I'd never want to travel again without my Cradlepoint PHS (kind of like the MiFi, but uses an external USB aircard)- it's just way too handy to be doing the multitasking-in-the-car thing like what's shown in that new Sprint commercial...
I'm pretty sure the older version of 568A (the standard itself, NOT to be confused with the T568A or T568B termination pair assignments) did mention this.
As mentioned in my post above, it's not optimal, but if you've already got wire in the walls and it's really hard/expensive to run new cable, this used to make sense. (It makes a lot less sense now with today's higher speeds, but is still applicable to many homes like mine, where running new wires involves lots of hole-cutting and wall and ceiling patching/painting, making it prohibitively expensive.)
I've never seen network electronics designed to work with such a setup, and as mentioned above, it really makes termination a bear, since you have to strip out the voice pairs and take them to a different set of punchdown blocks.
Although you *can* plug an RJ-11 into an RJ-45, every time I've ever seen this done, the outlets are terminated with two jacks: one RJ-45 (four-wires for Ethernet), and one RJ-11 for the middle voice pair.
It all works just fine, but there are good reasons you don't see it as a normative practice.
the idea of running a 48VDC pair that also uses a 100VAC ring signal right beside your ethernet pairs is scary
The 48V *DC* loop induces no noise at all, and in fact, since the positive side is at ground potential, it may actually help a tiny bit. 802.3af Power-over-Ethernet uses this exact same voltage, but puts it over either the "spare" pairs (there aren't any in GigE), or on the data pairs themselves - all PoE PD (powered device) implementations must support both. BTW, it's Ground and -48VDC so that the telco can put a sacrificial anode ground rod somewhere to eliminate corrosion problems - this isn't new, it's been that way for well over a century.)
Finally, the telco ring signal is a nominal 90VDC square wave, and I've designed wiring architectures for entire building campuses that used phone and Ethernet on pairs in the same jacket to avoid the huge cost of running new cable. I can honestly say I don't think I've ever seen the ring signal cause any significant data problems, unless there's something else wrong, too. Sure, it's best to separate voice and data if you can (mainly because of termination issues), but they work just fine together. (The worst problem is that the sides of the RJ-11 can sometimes spring the contacts for pins 1/2 and 7/8.)
Twisted Pair Ethernet is *incredibly* resilient - you actually *can* run it over rusty barbed wire - someone (SynOptics, maybe?) showed a cross-booth rusty barbed wire TP Ethernet link at Interop back in '93 or '94, IIRC. ('Course, that was back in the 10 Mbps days, with the old Manchester encoding, so it's about bulletproof...) Really, it's the same with any digital signal - I always laugh at people buying "Monster"-style HDMI cables - hell, it's a square wave: it either gets there or it doesn't, and expensive cabling will only make a difference if something is horribly wrong somewhere else!
In any case, if you use quality punchdown blocks and are careful not to blow the length or the dB budget for connections and terminations, it's hard to build an TP Enet that *won't * work, whether there are voice pairs in the same jacket or not...
P.S.: You're not really all that likely to feel the 48V unless you lick the wires, although it is enough to spark. Holding the wires when a ring signal comes through will get your attention, but that's true for any phone wiring...
Word is a word processor, not a text processor. There's a big difference: Word processors (Word, OOWriter, WordPerfect, AbiWord, etc.) mix up content and formatting, much like in the old pre-CSS HTML days. Formatting in WPs is generally specific text markup (bold, size, etc.) applied to a specific range. Text processors (most commonly the *TeX and *roff families) separate most formatting from the content - you just pour your text into the various content types and the text processor will apply whatever format it's been told. TPs tend to be more tag, style, or attribute-based - WPs with styles are somewhere in between, but as many have noted, actually using and enforcing correct use of styles or other predefined formatting in most WPs is difficult to impossible, especially with multiple editors/authors. TP documents are harder to get set up (TeX in particular is a PITA), but once that's done, it's much easier to focus on content and forget formatting.
Final thoughts: 1. Word isn't going away unless GoogleDocs or the like kills it, since it's still the only rich text editing tool of any consequence that is widely available, and it has the advantage of being sort-of compatible with itself. 2. Putting so much critical and nearly uneditable information in that little (normally) invisible paragraph mark at the end of your paragraph is simply insane!
Nobody's nailed the *real* reason most people choose PDF for document distribution: It's still the only foolproof way (for most people) to make sure that WIS (what **I** see) is WYG (what **YOU** get). If fonts and layout matter, then PDF lives simply because it can nail those down.
Word docs are awful for transferring documents, since if the recipient has another version of Word, or different/missing font, then all bets are off as to what will be at the bottom of page 8. (OpenOffice takes hits from both directions here, since it is both a varying version and its fonts are different.)
I *hate* the idea of PDFs, but I use them extensively, because the font industry's insistence on DRM to prevent font proliferation has screwed up documents for the last 20 years. Hell, we're only just now starting to *maybe* be able to use fonts and typography on the web for the first time, but it's far from certain that the foundries won't take steps to shut this down.
And what, exactly is really the difference between embedding a font in a PDF and embedding a font in a word processor document file? The former is legal, the latter is not. I guess it's that unbreakable Adobe Acrobat encryption or something... ;-)
There's no fundamental reason MS Word can't be a great program. All it needs is a pioneering visionary to thrash it down to a working core, to develop some well thought out guiding principles for how to organize the interface, to mercilessly eliminate the rampant bugs, to study how the current interface fails, and to rebuild it from that working core back up to a well-engineered product. But will that happen? Unlikely.
But in reality, as bad as Word's faults are, Word has seen far more of this sort of attention (especially with the new "ribbon" UI in Word2007) than any of its competitors. Although I think Word's UI is still light-years short of where it should be, it's also light-years *ahead* of it competitors, including OpenOffice, which from a UI perspective is a bad copy of Word's interface circa Word 4.3/95 with a dash of Word2000.
In fact, although I don't agree with many of the decisions MS made for the new Office ribbon interfaces, there's no question that they *really* did the work to completely re-think the interface. They even posted pretty detailed explanations on the web (Chan9, IIRC) describing how they reached the major decisions, and why the choice they made seemed to strike the best balance.
Also, keep in mind that MS is severely hampered in what it can do by its user base, who are extremely sensitive to change of any kind (inertia, training costs, institutional stupidity, macros, etc.), even if that change results in a far better product.
And Microsoft is at least *trying* to think about user experience and common look and feel across its applications, something that OO seems to be relatively incapable of - screaming about the UI dates back to the StarDivision days. It's not that OO doesn't suck, it's just that it sucks less than any other free alternative to Word. (WordPerfect also has its flaws, but it absolutely kicks Word's butt for stability in producing really large, complex documents, which is why it's been the legal industry's favorite WP until very recently.)
Thanks, I didn't know the Russians were keeping their modules up there. Serves me right for not RTFA...
Regardless, I'd be really surprised if some variant of the evaluation I described didn't enter into NASA's thought and decision process.
Give the money in well structured grants to the private sector, like Burt Ruttan and Elon Musk, at least they are smaller, leaner and willing to think outside the box
Actually, those two guys may be the biggest reason NASA wants to bring down the entirety of a station it only barely owns half of in the first place!
Think this out a bit:
1. Once the shuttle retires, NASA will have no manned spaceflight capabilities to speak of, which the ISS requires to stay up. ;-))
2. Because of 1, NASA would have to "abandon" the ISS by leaving it unmanned.
3. Entrepreneurs and inventors really love a challenge, and a prize. And the ISS is quite a prize.
4. Getting someone aboard the ISS may well be legally "taking possession" of it. (I'm making the plausible assumption that the salvage laws in space would be found to be the same as, or largely similar to, those of the seas.)
5. The ISS cabal definitely doesn't want a spaceborne Sealand, and they'd rather torch a half trillion taxpayer bucks than let that even be a possibility. Things get worse if you contemplate unfriendly countries occupying the ISS, possibly as an orbiting recon and weapons platform. (NORK-ISS, anyone?
This may be a far more credible rationale for "de-orbit" than any bogus "safety" argument: NASA and the other ISS owners can't keep occupying it, which is the only thing that perpetuates "ownership", and they can't stand the thought of anyone else getting to own it, so they'd rather destroy it.
Destruction may even be sound policy, depending on the actual strategic risk, although once again, the taxpayers whose wealth was confiscated for this boondoggle get screwed. I love space technology, but tend to agree with Walter McDougall that a huge unintended consequence of the space race was to destroy America's private innovation and set us on the road to big government control of our lives.
But it's interesting that MS is making Windows 7 hard to use on netbooks and older laptops don't you think? That's because of something I haven't even seen anyone mention - the Win 7 RC is *only* available on DVD. This has been enough to deter me from trying it, since every single machine I would want to try it on does not have a DVD drive. This also says something about the bloat of Win7 as opposed to XP - I'd really like for the MS folks to take a serious look at Puppy Linux to see how you can build a small, light, fast, highly functional OS that fits in just a couple of hundred MB, even with a rich app set. (XP Puppy - yeah!)
Heck, for that matter, I don't even own a DVD burner, so I couldn't write the disc here at home even if the PCs and laptops I'd like to try it on could read a DVD. This is simply because movies are the absolute last thing I ever want to do on a computer, and I stopped using optical for backups back when the size of a CD was adequate for most things.
--
I am becoming gerund, destroyer of verbs.
THAT is an excellent sig, my friend - well done.
Good point. I agree Tufte has many acolytes who simply parrot whatever he says, and that much of what is says actually *is* simply a pronouncement of his opinion, often without sufficient analytical thought to back it up.
That said, Tufte's a good writer, a decent thinker, and has been right often enough and on enough topics to have built a certain degree of credibility. Personally, I think "The Visual Display of Quantitative Information" is indeed a great nail on which to hang a career, but his ideas on user interfaces and the like are considerably less well-thought-out (especially sparklines, which are useful in some cases, but certainly not the cure-all he seems to think).
FWIW, I have never taken any of his courses, and don't even own any of his books, although Visual Display is on my wanted list.