Slashdot Mirror

...which is exactly the same as the second edition. Almost. But not quite as painfully obvious a reprint as the OpenGL book that usually accompanies it (the Red Book). It just had some warnings that OpenGL 3 is, like, totally different, and saying that we are not going to bother with any of that, despite announcing that prominently on the cover of the book. Nothing on, for example, how to do transforms without all the old pipeline commands. Lots on, say, display lists, instead. Display lists that have been unofficially deprecated now for about a decade. Oh, well.

And for those not in the know, the Orange Book and the Red Book together form the unofficially official documentation for OpenGL.

Someone really needs to step up to produce quality documentation for the new lean and mean non-backward-compatible profile of OpenGL 3 if that is to make any headway. Or maybe that lets-make-the-API-even-more-low-level approach was wrong to begin with, and people are just afraid to say that out aloud like Mark Kilgard of NVIDIA recently did (see slide 35).

Anyway, if you are going to work with OpenGL, you need to read those two books. But you can just as well buy the previous edition.

The problem is that most Windows programs don't have any installation method other than the installer, so if Windows decides the installer needs admin rights, there's usually no way to do it without them.

Most Windows programs I've installed did not use the Windows installer, nor did they include an uninstaller. I don't know how many tymes I had to wipe my disks and do a clean install of Windows, in part because a program left or altered keys in the registry. Using Norton System Works to uninstall didn't always help. On the other hand every tyme I installed programs in Linux I had to log in as admin and on the Mac I'm typing this on now even when I'd logged into the admin account I still have to enter the password to install software. And yes, I've owned and used all three OSes.

here's no reason to prohibit the user from installing software locally to their home directory

Viruses, crapware, and spyware are very good reasons to require admin privileges instead of allowing users to install software. As I said earlier employers are even disabling or removing CD/DVD drives and USB flash drives.

Falcon

CouchDB

.

Perl 5 is developing quite nicely. And number of users is raising (in absolute numbers, and recently, even in relative). See http://www.slideshare.net/Tim.Bunce/perl-myths-200909

I have not used Livescribe recommended by another poster. But it seems to me you will not be able to record your audio unless you actually walk to the blackboard with your tablet. Of course if you are always in tablet mic range then it might be okay.

Another thing is quality. You can get a pin mic (see audio technica brand or there are others) and voice recorder (they all have sd cards and usb these days, the most popular one for business will record 1000 hours or less at MP3 quality, best I am told is Olympus but there are other brands and some are pretty small).

This might give you better quality and you can pin the mic to your lapel or shirt front while slipping the recorder in your pocket. But if you can indeed just use tablet for all drawings and always be near it you might have good enough quality plus be able to synchronize. Of course this assumes you have a way for the student to play it back. I think personally it would probably be much better quality if you do this:

Create written course notes in advance with nicely written equations etc. on your tablet
Draw on blackboard with tablet and record that, give this and the advance notes to the students on your course website.
Use voice recorder to get high quality sound and dump the sd card to your website. Verbally say which page of notes, or blackboard drawing, you are talking about.
Optionally have a video camera synched to voice recorder, depends on how animated you are and how important are the gestures at the blackboard. Maybe not so important.
Then organize the files on your site, with a wiki or blog and maybe a rss feed to let people know when it is updated. Maybe scribd would be useful too.
I saw some info at
http://www.slideshare.net/hebertm3308/interactive-classrooms-presentation
and http://www.kn.pacbell.com/wired/fil/pages/listweb20s.html

A couple of months ago, I became aware of the concept of single-tasking. Back then, the presenter claimed the philosophy was so recent that it didn't even have a Wikipedia page. (It still doesn't.)

The idea behind single-tasking is the same as in this article: multi-tasking gives us the feeling we're doing a lot of work, but actually we lose a lot of time mentally switching between different "contexts". Working on a single task until the end, and then taking on the next one is better.

in Ohio at all.
80% of our electricity is coal generated.

Apparently Ohio's potential wind power is pretty good, onshore as well as offshore. According to one person mentioned in the second link above Lake Erie along the Ohio shore can provide more than 100% of the states electrical needs. It may not be enough with a lot of EVs on the road but to discount EVs in Ohio just because most of the state gets it's electricity from coal doesn't work.

Falcon

A recent proposal from the UC Santa Cruz EIS lab (also mentioned in the article), an Automated Game Designer: http://www.slideshare.net/rndmcnlly/the-intelligent-game-design-game-design-as-a-new-domain-for-automated-discovery-1784151 It's not about making a bot that can behave intelligently/interestingly in a restaurant setting... what are the broad applications of that? (As other people have pointed out, the bots may come out pretty demented and flavored like The Internet.) It's about making a game designer that can design games on its own, learn from its own experience, get MINIMAL human input (not 10,000 plays online). The computer designer can do what the computer is good at (enumerate all possible play traces, look for instances of accessibility/cheats/funky behavior the designer might not have intended or expected) and the humans on the side can do what they are good at (shaping, polishing, collecting a few human play traces).

Sadly I couldn't find any connection able to upload the 22Mb audio file of the keynote, but I'll try again tomorrow.

In the meantime you can check the slides at http://www.slideshare.net/qgil/maemo-harmattan-qt-and-more

Quim Gil

Is it really necessary to write a 200+ page book on how to write 140-character "microblog" posts?

Agreed. After reading the preview, it's basically a tutorial on Twitter. From the link to the preview I found these general sections:

1. Get Started
2. Listen In
3. Hold Great Conversations
4. Share Information and Ideas
5. Reveal Yourself
6. Twitter for Business: Special Considerations and Ideas

I'm sorry this looks like a "how to use Twitter" book that could, at best, be a social networking book. What is O'Reilly doing publishing this book?

The review tends to center on formatting and typesetting. Great. If I want to know how to publish a guide on using my site, I understand this book sets the bar.

But how does it get a 9 out of 10? Did the reviewer walk away a better Twitterer? What need is there to buy this book if you could just save the time and money by diving in or reading other Twitter posts?

The difference, of course, being that design or operation errors in wind power systems are highly unlikely to kill dozens of people instantly, hundreds or thousands of people over months and years, nor render thousands of square miles uninhabitable for sufficient multiples of the half-life of uranium, plutonium, or whatever other radioactive by products are released into the water or atmosphere in an industrial accident so as to avoid a legacy of genetic deformity in the affected area. See: Chernobyl Legacy: a photo essay by Paul Fusco.

I'm a proponent of nuclear energy in the abstract sense that it would be useful to help us out of the acidified ocean mess we're heading into, but there exist enormous problems which the nuclear power industry didn't fully appreciate in the 1970s (when the technologies in all currently operating reactors were designed). The industry will continue to treat waste disposal and many components of accident risk as externalized costs, left to their own devices. (Externalized costs are a natural by-product of a capitalist system. To counter balance that effect we invented the useful concept of government regulation. Note that quite similar externalities also existed in the communist systems which produced Chernobyl, which, like the capitalist systems of the west, also didn't do natural resource accounting nor assign a value to the cost of safety, accidents, pollution, nor maintenance of a thousand-square-mile "exclusion zone" for hundreds of years, in their economic models and business plans.)

Much safer designs are possible, but we need dramatically safer, accident-proof, terrorist-proof, designs. We can probably do that, too, but it will cost more, and we'll need to actually do it, before we deploy a new generation of plants. Oh, and uranium and plutonium are fossil fuels, too. Here's a discussion of Thorium, molten salt, low pressure containment, reactors.

I should correct myself, Search Server 2008 Express scales upto 400,000 documents not 1,000,000 primarily due to 4GB limitation with SQL Server 2005 Express. If you have SQL 2008 Express, I'd have to check the scaling.

One of the big benefits with Microsoft is the ability for granular search tuning. Enterprise Search is a very different from internet search and having access to the search algorithm is key to get better results. Below is a partner who deals with GSA and SharePoint/Search Server.

http://www.nonlinearcreations.com/blog/index.php/2008/06/30/google-search-appliance-and-microsoft-search-side-by-side/

If you want my powerpoint presentation on Search Server 2008, please visit:
http://www.slideshare.net/ukdpe/microsoft-search-server-2008-technical-overview

i just wonder if they're as desperate as the couchdb bozos: http://www.slideshare.net/mattetti/couchdb-perform-like-a-pr0n-star

Ah, the infamous "common knowledge". I will tell you that any of Sun's recent SEC filings are pretty pathetic, and SPARC has done so well that the company is for sale. That should tell you something.

Maybe you don't know how to Google so let me help you out. http://www.slideshare.net/earningsreport/sun-microsystems-q2-2009-earnings-releases there's a slide with a breakdown of systems billings by category.

SPARC ENT: 578mln
SPARC CMT: 338mln
x64: 176mln
Other: 156mln

That's 916mln for Sparc based servers

Hardware/OS Support billings were around $900mln for the quarter. When you consider most of the hardware sold was for Sparc servers (Storage was only around $500mln), it's safe to assume that most of the hardware/OS support was related to SPARC servers.

That's an argument by precedent: "The sun has risen every morning, therefore it will always rise."

CEOs should be forward looking, not backwards looking. Does SPARC really stand a chance against the x86 world at this point, especially as the x86 world is beginning to embrace PLD and GPU computing as well?

No. It's a statement of fact. Ellison tried to be forward-looking back in 2002 when he decided to switch the Oracle developers workstations from Solaris to Linux. He expected Linux to be the top deployment platform for Oracle. Because if you're smart, you develop your application on the most popular system your customers use.

Here we are seven years later and that hasn't happened. Either he was too forward looking, or he was wrong. Both are bad. CEOs aren't supposed to take big leaps with a company that large. They need to take big steps, with one leg firmly planted in the present.

As for open source databases, I use them a lot. In the past couple of years, only PostgreSQL and I have been deploying it on Solaris/x86 on whitebox servers with great results.

When I've deployed Oracle, it's been on Solaris/SPARC, but I know some big companies that are deploying on Linux as well.

I don't know the exact numbers of Oracle vs DB2 on AIX, but I wouldn't be surprised if it was close to evenly split. If Oracle wasn't the primary DB for HP servers is likely MSSQL, unless you're talking only their unix servers in which case I would be surprised if it's not Oracle. In that space, the only real choices are Oracle and DB2 and you would expect HP to not want to have their customers give money to their biggest competitor.

I don't understand the general animosity towards Sun around here. Generally people tend to root for the underdog, and I believe that to be especially true in the geek world.

Sun's biggest competitors are IBM and HP. In that instance, Sun is clearly David. But people want to pit Solaris vs Linux and in that case Sun is Goliath in terms of company size and revenues.

I think it's unfortunate because of all the work Sun did with open source.

But if one starts thinking about it, it has the ironic effect that TPB is a driving force of consolidating the market power of the major labels rather than driving forward any new music. The conclusion has to be that "pirates" are just as little resistant to the major label marketing as any other person. Even though there are thousands and thousands of artists out there that want their music to be shared and listened to, they are widely and effectively ignored by the masses. In fact, one might say that TPB and the likes are countering the development of new markets, simply because the gap between the heavily marketed music and 'the others' is wider than ever, when the bare naked truth about peoples taste in music is put into such a system.

Indeed. So obvious, so seldom stated. I made it point 0 of http://www.slideshare.net/mlinksva/five-myths-about-the-future-of-culture-and-the-commons-presentation

The core service platform at LinkedIn runs on Java

They do have a Facebook application that runs on Ruby.

There's no need to create new licenses to have CC-like easy-to-understand software licenses. CC has experimented with "human readable" deeds for a few software licenses and could work more with groups like FSF and OSI to do more and improve on those.

Noncommercial public licensing failed in software for good reasons, and it would be really dumb to introduce it at this point. Many people complain about NC culture licenses, but for software, they are much worse for a variety of reasons that I'll write about eventually, but see some of the bullets at http://www.slideshare.net/mlinksva/how-far-behind-free-software-is-free-culture-presentation

There are lots of poor software licenses out there, but the current generation of ones that are widely used and had a ton of attention during drafting are excellent, ie Apache2 and A/L/GPL3. To the extent they are long it is because they need to be (excepting preambles perhaps). CC licenses are also pretty long.

Running on Rails has forced us to deal with scaling issues - issues that any growing site eventually contends with - far sooner than I think we would on another framework.

That is probably true. However, I would count that as an advantage -- better to deal with them sooner than later.

At this point in time there's no facility in Rails to talk to more than one database at a time.

There are many, many ways to talk to more than one database in Rails. In fact, it is possible to swap out the entire database layer of Rails and use another ORM, or no ORM at all. On the bleeding edge -- and Twitter might actually be a good candidate for this -- people have wired up Rails to CouchDB, which provides trivially scalable multimaster replication, and which, being HTTP, can be thrown behind any old load balancer -- which brings this back to a "just throw hardware at it" problem.

All the convenience methods and syntactical sugar that makes Rails such a pleasure for coders ends up being absolutely punishing, performance-wise.

Some of them do -- a good example would be Symbol.to_proc.

However, Merb proves that this is not actually a Ruby problem, it is a Rails problem. And Rails and Merb are merging some point in the near future.

It's also worth mentioning that there shouldn't be doubt in anybody's mind at this point that Ruby itself is slow. [...] I think it's worth being frank that this isn't one of those relativistic language issues. Ruby is slow.

Somewhat true -- after all, Ruby 1.9.1 did double the performance of the language.

But, relative to what?

Turns out that, at least compared to other languages and frameworks (like PHP), Ruby is not slow.

It's also worth mentioning that while all of the Twitter alternatives may have enjoyed better uptime, they haven't had nearly the amount of traffic that Twitter does. We don't really know if they can scale -- but even supposing they can, Twitter was there first. And while they complain about those nice features being slow, they probably owe their success to those features for getting their product out the door faster than their competitors.

It's also worth mentioning that this interview is almost two years old. Rails changes a lot in two years. In fact, Twitter were early adopters -- two years before that interview, Rails had only just shared commit rights. Two years before that, it didn't exist at all.

It might be worth asking what version of Rails Twitter is using, and if they've noticed a change since then.

The unequal status exists only for the original work, except in cases of assignment. Modified works with many contributors under a NC license -- indeed, nobody can ever profit (modulo fair use of course :)). At least for works with many contributors, NC creates a commercial anticommons. It's possible this is more damaging for software than non-software, at least partially explaining why noncommercial licensing is essentially nonexistent for software. See slides 15-27 of http://www.slideshare.net/mlinksva/how-far-behind-free-software-is-free-culture-presentation for some handwaving on this.

You do realize neither Ruby nor Rails is an acronym, right? Ok.

The only argument for it in the past for web dev is RAILs and there are plenty of MVC frameworks for PHP now including PHPulse, Cake and Codeigniter.

And those don't compare well, even to Rails, certainly not to Merb.

And Merb is going to be merged into Rails.

under large loads, it buckled

Hardware is cheap. Couldn't you throw more at the problem?

On a side note, I will use PHP on my servers before touching Ruby since I see no advantages for using it over PHP.

If you see no advantages to using it over PHP, you obviously haven't looked very hard.

Off the top of my head: Ruby has better syntax, a better object model, runs faster (really), better standard libraries -- Rails aside, Ruby tends not to pollute the global namespace with bullshit like mysql_escape_magic_quotes_no_really_I_mean_it_this_time...

PHP's advantage? Lots of unimaginative programmers like you know it, and it's slightly better at mixing code and data, since it's really just a Turing-complete template language. But I like haml anyway, and regardless, the template is one tiny part of your app -- the actual application logic should be buried in the model somewhere.

So for the majority of the program, PHP has no advantage. For the tiny minority where it would, in a well-designed app, haml makes it look so ugly it's not even funny.

You know, you have a point. I think all dynamic languages could probably learn something from the recent Javascript speed-up projects...

Except Ruby isn't slow.

And yes, I am happy about it being twice as fast, because it was already fast enough.

The worst mileage cars in our country today (generally) are OLD clunkers which pollute like crazy (manufactured before 1982) and SUVs, which are mostly built on truck frames.

Getting as many of the first off the road is in our general interest. It reduces asthma and generally cleans the air. One of these cars produces as many as (pulling a number from the air) 20 modern cars. It's beneficial to society. Does it benefit $5k worth? I'm not sure. Very possibly, given that there are economy stimulus aspects to this project which somewhat lower that cost.

Why 1982? Because emissions haven't gone down since then, which is pathetic. http://www.slideshare.net/marcus.bowman.slides/vehicle-technology-improvement-curve-462599

Getting as many of the truck frame SUVs off the road is another benefit to society. They kill people in numbers much greater than real cars. Both the drivers and the people they hit. We pay for those accidents via higher insurance costs.

If the $5k is for replacing either of these with a modern, non truck frame vehicle, it's probably worth it. If we're replacing a 1992 SUV with a slightly smaller one, it's not benefiting anyone outside of the auto industry very much.

Couple of points:

Of course, caching and setting up distributed workload processing queues will be necessary in any framework, but that's up to the development team working on a specific app rather than the framework itself.

True -- but Rails does make caching easy.

You may want to look at how Rails 3.0 (the next release of the Rails framework) is incorporating several ideas from Merb to make the framework lighter-weight and more extensible

It's more than that -- Rails 3.0 is also Merb 2.0. So, Rails 3.0 should gain all of the advantages of Merb -- one of which is raw performance.

For offline processing in a Rails app, look at plugins like workling/starling and other message queues.

For what it's worth, Merb has a simple threaded backgrounder built in.

HERE ARE SOME SLIDES FROM LIMITS TO GROWTH that I've uploaded. They concern only scenario#2, which is but one of the scenarios developed in the model (and the one I think is turning out eerily close to reality).

Slides 11 and 12 are particular sinister to me.

Obviously, I'm placing them here totally out of context, but when you read the book you see that they do make sense, and how these global variables feedback into each other. (Note. Other slides loosely related)

Let's see. You'll get a CS degree but don't feel like writing code for a living. That's a tough one.

Are you a "people" person? All those introverted geeks need to talk to each other, make decisions and agree on stuff. Something that they (on average) do very poorly. You would have a career in product marketing, since you understand the geeks and can talk to them.

If that makes sense to you, then short-term, your best bet is to join an open source project and volunteer to *organize* stuff. Not code, but organize. You'd be amazed how badly needed it is for most projects.

--
the elephant in the room: How to Make Money with Open Source?

I really wish the motion would pass. Finally, we could extract soundbites from the RIAA's lawyers to show how ridiculous their position is.

But my guess is that it's not going to happen: it's a long shot. Allowing media in the courtroom is the exception, not the rule. What I wish for, I usually don't get...

15 years ago, I used to buy CDs. I couldn't listen to the tracks ahead of time, often 90% of the album sucked. But I had to pay the $15 anyway. Now I buy my music legally, online, but I often just buy one song (99 cents), the ones I really like.

Guess what, the RIAA's business is dying. They don't provide value anymore (if they ever did).

When that happens to a corporation in America, you have two options: Change your business model, adapt and become competitive again.

Or ask the government for a bailout. Dear RIAA, stop the lawsuits, just ask Uncle Sam for $100 billions. It's much easier and faster than your current approach.

--
Free and Fair, Friend or Foe?

You're absolutely right that Ruby is slow, and Ruby 2.0 will make it faster. But relative to what?

It turns out, someone has done benchmarks, and the results may surprise you. It turns out, Ruby really isn't that slow.

As for the ease of deploying, you can do it seamlessly with mod_passenger nowadays

You know, I really don't get this -- how, exactly, is Passenger easier to deploy?

For that matter, how is Rails hard to deploy? I've always found Capistrano to be much easier (and more powerful!) than sending PHP files over FTP.

Yes you are right but Google/Amazon are read-heavy so a big hash table works perfect since there is less disk access - however FaceBook is both READ and WRITE heavy so its more challenging. From their last SIGMOD Cassandra paper, I gather they use some sort of in-memory write which is then synced with their bigtable. Very cool stuff!

Just because it's written in Scala doesn't mean it's shitty and unmaintable! Just the latter.

(Talk slides, if you're interested.)

Start by reading Stop death by Powerpoint. What it says is applicable to any presentation in the world. If you dont' have a cannon for projecting stuff, maybe you can type something on video, or even bring a laptop and attach it to a tv. Or print something. But bring in some visuals. And practice; if you don't have experience making presentations, practice. Children get bored more easily than adults.

So, that's *how* I would explain it. Regarding *what* to explain, it doesn't make sense that you go into detail. Your day to day probably is too abstract for them to pick any sense, mainly because they probably don't know anything about aerospace anyway.

I would put myself in context. I'd start saying that aerospace is very complex. And complex means lots of things to do, which means lots of people doing a lot of things. And those tasks depend on each other (I'd probably use the metaphore of a body, or show them a picture of a plane, with the aprox amount of people needed for making a wing, the tail, etc, from start to end). I'd explain very rougthly, like with one phrase, what are the main tasks for building a plane. And then explain how did i fit in there.

Then, i would explain how a regular day is. I'd try to relate it with things that they might know ("how many of you use messenger? xbox live? well, the server on the other side, i take care of it)

I'd explain what happens when things go wrong (show explosions! put a clip of Rambo shooting rockets and say "this is me turning off the server that day"!!). Try to think of some anecdote that happened to you and they might find interesting.

Just my 2 cents :)

According to the published records, the Perl Foundation awarded at least $200K in grants for Perl 6 development during 2000-2006.

That could buy two years of a modestly-paid software developer's time (figure in self-employment tax, health care, et cetera). TPF crammed two funded developer years into six calendar years for Perl 6, and it's still not done. (We might be up to four funded developer years for the specification, the test suite, a virtual machine, and the language implementation.)

Again, the distinction between stable, shipping products and endlessly evolving projects seems to elude you.

No one has claimed that Parrot or Perl 6 is a "stable, shipping product" right now. If you really want to claim that this makes the entire project "vaporware", please feel free to apply that moniker to Ruby 2, PHP 6, Python 3.0, GHC 6.10, Java 7, ECMAScript 3.1, and a few hundred thousand other projects in progress -- at least, if you're at all interested in applying words fairly.

While the Perl 6 project may have attracted some of this funding, and some of the Perl 6 work was backported to Perl 5, I very much doubt that this adds up to being a net benefit for Perl 5.

You may find it interesting to hear that the reason Perl 5 (for one example) has orders of magnitude more core tests than Ruby or Python (and let's not mention how much the culture of testing has permeated the CPAN) is a direct result of the Perl 6 project. See slide 43 in Tim Bunce's Perl Myths 200802. (If you don't believe him, believe me, because I was there, I wrote a significant amount of them, and I did the research to which Tim alludes.)

Also, I'd love to see your list of widely-available, cross-platform C99-conforming compilers -- as someone who writes a fair amount of C code to run across several operating system and compiler combinations, I can think of several vendors who can't seem to support a decade-old C standard.

Except of course...

- Perl is not dying, in fact see this presentation about how well it actually does in terms of Perl jobs vs Ruby or Python.

- Says who? I find Perl syntactically beautiful and _consistent_. Much more so than php or ruby. Python is nice, but whitespace sensitivity is a burden some people are not willing to bear.

- While I certainly respect personal decisions as to which tool is the best for a given job, I personally would be hard presssed to find a better language for web development. That is even without talking about CPAN - because with CPAN Perl wins with a landslide.

- It's easy to write bad code for those who don't have the willingness to learn the language basics. Or to answer you with a quote: "If you want to shoot yourself in the foot, Perl will give you ten bullets and a laser scope, then stand by and cheer you on". This doesn't mean however that the language doesn't lend itself to good practices! There are reasonable community standards that are well known to anyone even tangentially involved with Perl. Perl let's you write code that is trully messed up, but only if the programmer writing it deliberately formulates it so. If a programmer decides it's a good idea to violate all reasonable coding standards and write code in a mess of a fashion, then that's the incompetence of the programmer. True, Perl doesn't actually FORCE you to avoid that kind of behaviour, but if you don't avoid it by yourself, then you have no idea about programming anyway. Comparing this to java, I'd rather be able to tell by glancing at a programmer's code in Perl that he's no idea what he's doing than to use java like strictness where source code looks like it's from a cookie cutter and discover that a programmer is lousy based on business code. This whole "Perl is unreadable wah wah wah" is just another tired chestnut, sorry.

While you're at it, why not just say that Apache is dieing too, based on that logic? The fact is, if you look at the Perl job market, it's booming.

1. Rush, a ruby shell. Rush strikes me as a very cool project. This slideshow is a good introduction.
2. iPython with the "sh" profile. About halfway into this article they discuss it.
3. Hotwire, an "object-oriented hypershell"

Those people you're protecting from chemical weapons from 1980. What did they do to deserve this:

http://www.slideshare.net/Peety/baghdad-before-after

Suggesting that we toss Israel into the mix is irresponsible and typical war hawk doublethink.

I'm sad for humanity and slashdot right now.

Maybe someday we'll find our way again :-(

http://www.slideshare.net/linhares/outline-of-globalization-course-at-fgvebape Check out slide 9, which compares the explosion of engineering degrees in China, India (& to a certain extent the EU) to the US and Japan. I use it on my classes, and people think it must be bogus. Data from Morgan Stanley, by the way.

Demand shaping enabled by smart meters is happening today (check out Florida Power & Lights programme) and works. Tricks like turning a water heater off for 40 minutes during the middle of a summer's day when the air con load maxes out can deliver major savings.

The real problem is that the entire energy system is demand driven: when you turn the a light there better be electrons there or we have a brown out. Want we need to do is convert the energy system to supply following: devices use power when it is available.

The trick is the feed the price from the energy markets into the home where we can run an internal energy market. Your appliances bid for the power they need and the highest bidders win. Our smart homes and the appliances themselves decide when is the best time to soak up excess energy supply. Lights will always want power, your water heater might not if it's hot, and your washing machine might be happy to wait until the power is cheap late at night.

This also nicely accommodates your own generation (solar cells, CHAP ...)--give it a price and add it to your home energy market. You can even sell you excess back onto the grid, if it makes sense.

Like all increasingly scare resources, we need to use less. Technologies like desalination are only making it worse (desalination needs a lot of energy, and we need more water to make more energy). If we want a sustainable energy system then we need to be a lot smarter about how we generate and use electricity.

A little more detail at SlideShare.

r.

PEG

A similar thing happened with FreeBSD 4->5. FreeBSD 4 was at some point the fastest, most stable and most administratively sane operating system for the x86. Linux 2.4 was already ahead in SMP scalability, but since this did not affect many users back then, not many people cared.

Then FreeBSD 5 was planned to include a major architecture shift to modern parallel programming, which required changing almost all of the kernel code sooner or later. FreeBSD 5 was downright unusuable until the 5.3 "stable" release, which was tolerable but still performed a lot *worse* than FreeBSD 4 for many workloads because it was half way between the very well optimized uniprocessor code and the very raw and experimental multiprocessor code.

FreeBSD 5 was looking so bad even early on, that Matthew Dillon predicted (incorrectly) that it would become unmaintainable and fail to modernize to new usages. He forked FreeBSD 4.8 into DragonFly BSD, and developed it with more clever, innovative and forward-thinking designs.

However, FreeBSD 6 and finally FreeBSD 7 polished things up to the point that a slightly reconfigured FreeBSD 7.0-rc1 will compete with and often outperform Linux 2.6.22 for database throughput, with comparable responsiveness to CFS in 2.6.24. FreeBSD 7 is stable and fast, and still includes as many features as the old FreeBSDs did, including running well on old hardware. FreeBSD 7 humiliates the current DragonFly (and in fact, all of the BSDs) in throughput and scalability alike. Here, have some numbers: http://www.slideshare.net/sim303/7020-preview/

An exceptionally bad 5.x branch did not kill FreeBSD. So I don't suppose KDE 4.0 will kill KDE either, or even significantly reduce its mindshare. Especially if 4.1 comes out by the end of the year and improves significantly, KDE might gain even more adoption than it has now in 3.5.

I have no comment on IE or Windows explorer, but the Office 2007 UI was not changed just for the sake of change. It was changed because feature rich applications like Word, Excel and PowerPoint had already passed the practical limits of a menu/toolbar and dialog based UI. Users had to dig through long, often confusing menus to find an entry point to an often complicated dialog. Many users were faced with an overload of options and simply gave up searching. A main goal of the Office UI overhaul was to improve discoverability (resulting in the ribbon) and to make it easier to format documents nicely (resulting in galleries). This presentation by Jensen Harris pretty much sums up the reasoning behind the UI change.

And the reason it seems like Office, Messenger and WMP are made by completely different software vendors is because basically they are. Microsoft does not have some monolithic development staff that rotates around from project to project ensuring consistency. It's a very large company with many different groups of people working on specific products.

This utopian society may soon be reality for most of us.

I have news for these people, internet connections go down, servers crash, on-line servers get hacked. Laptops get stolen, hard disk drives break down, viruses can destroy your data... Data on some online provider's disks is generally better protected, the real danger is that provider or the company hosting it might go broke / rogue / ...

I will always buy standalone software. You can pry my copy of Office/Visual Studio from my cold dead hands, or when I sell it for say 50%. Take that away, and I can use OpenOffice, and good old GCC/G++.

How many times have you had to pay for an upgrade that you needed because it fixed some bugs or because your old version didn't support your new operating system anymore? Software vendors are always in it for the cash, whether they sell you something (perpetually unfinished) or rent you a service. Judge them by the easy with which you can take your stuff elsewhere (try reusing Word documents elsewhere without problems).

Personally, I don't use any of those web 2.0 online tools (although I do look at online presentations occasionally - e.g. this), but I realize that they do have their use for the perpetually connected among us. I would never use such tools if I couldn't make a local copy of all my data easily though (poor webmail users...), or if I had to deal with highly confidential stuff.

The student was fully correct to 1) try to negotiate with the teach and 2) when that failed, switch to a more competent one. If the teacher is *requiring* a format that can be used by only one application on only one platform (both of which are expensive to acquire, operate and maintain) then they have too much ignorance or too much of an axe to grind to be allowed to continue teaching. To add to the damage, that application munges older files in older formats

'.doc' is a whole shitload of different formats, some very differentm some only a little different. However, it is because of the differences that sales for new versions of MS Office are driven. If the old programs could read the new formats, then we wouldn't have that problem. Why else do you think that MS Offfice 2007 munges your old files?

If MS published the specs for the old binary formats, we wouldn't ahve that problem either. Or if MS Office supported an open format like OpenDocument we wouldn't have that problem.

The way off the treadmill is openformats even for MS Office.

If you and your friend are so bright you could go into the options and set Office 2007 to save in Office 97-2003 compatability mode.

What part of "he can select an earlier format, but then it saves as read-only" did you not understand? Office 97-2003 compatibility mode has three different meanings in Excel, PowerPoint, and Word, and the one in Excel prevents documents with "new features" from being saved to a file that can be edited by a previous version of Excel.

> That's clever. Did you miss the part of the standard where that sort of thing is required for backwards compatibility? Apparently only Microsoft cares about that sort of thing, so that's why it's in the damn standard.

*REQUIRED* for backwards compatibility? What's wrong with converting them to standard markup instead of hacking in undocumented things like that? And besides, you do realize that that means that every word processor from now on has to know and care how ancient word processors once worked, simply because the newly converted documents are really old documents in a shiny OOXML wrapper.

If they really care so much about backwards compatibility, shouldn't they bother to document how these things work? You know, so that everyone can use them? Oh, right, they don't want anyone else to use them...

> If the ODF fanboys and FSF-sponsored trolls don't care about that sort of thing, I reckon they can safely ignore them and not implement them. But I guess that's no fun because it eliminates one of the fav memes being thrown around to prove that the standard is somehow deficient.

Oooh! I had no idea the FSF was sponsoring me! Where do I get a paycheck? Oh, right, I don't. It's really clever to argue like that, too, when a "rogue" Microsoft employee was caught trying to bribe Microsoft partners to join their NBs and vote in the ISO. True, we can suppose that that one employee stepped over a line, but you do realize just how many Microsoft partners have jumped out of the woodwork to vote on this, right?

If "we" are the "trolls", why are the OOXML supporters the ones with financial and business incentives to do things that aren't sensible in terms of technology or standards? And I say "we" loosely because I'm just some lone geek nobody who knows that he will have to support the crap that is in MS Office at work and hates it because of that. I'm an absolute nobody that no one in either camp has heard of, but I gather and submit most of the ODF articles here, including Slashdot story on those back compatibility hacks that you're trying to debunk (which was my favorite, BTW).

> Complete with "OMFG the 1.0 implementation as a BUG!!! Therefore the standard SUCKS!!!"

Funny that: why is it only one side that brings up technical arguments and actual data on what actually works and what doesn't, while the other side calls people names? Oh well, I really hope you never learn that lesson. If PR actually had a clue, they might be more dangerous in spreading FUD to geeks. Though I suspect that you were just trolling me and aren't actually paid by anyone, I figured it was worth posting for the people who are actually interested in this information.

Well, you see, the reason I hate all those old bugs is because Microsoft shows no interest in fixing them. ODF is going through revisions to actually *fix* all of its problems. OOXML? They have tons of people spouting "it's not THAT bit a deal! calm down!" and NOBODY actually fixing the damn thing. Hint: if you actually *fix* the problems identified in the NBs' votes, the "no with comments" votes would become yes votes and you wouldn't need to cram the committees full of Microsoft Certified Gold Partners.

Oh, and while I already explained why "just don't implement" them isn't an option (those legacy tags NEVER go away!) I might as well point out that Microsoft does a really crappy job on backwards compatibility in Office 2007. You know, their flagship product?

Might want to know about THAT before you go shooting your mouth off... Or not. Frankly, if you don't, it just makes things a lot easier for me.

I was listening to a show on CBC radio (gov't-funded NPR-like radio in Canada) a month or so back and they had a marketing guy talking about the value of brands. The speaker asserted that even bad brands have tremendous value, because they need to be focused, not established. Establishing a brand takes years and a shit-pile of money, with no guarantees, said he. From this guy's perspective, there is nothing more difficult in marketing and sales than establishing a brand, where a brand is a gut feeling about products+prior experience+what you've heard+service+etc. It's all that stuff that is evoked when you hear the company name, see the logo, think about buying a product.

This is completely off my cuff, but I think Linksys is a very established brand in residential markets, where 'Cisco' isn't. My girlfriend's son (first marriage stuff) even called his wireless router 'the linksys' last week ... and his wireless router is labeled by Dlink.

He sure as shit didn't call it 'my cisco'.

I call this move a mistake. Here's a Slideshare doc I cam across a few months back; the writer can't spell 'Porsche' correctly, but nonetheless I think it's a good intro blurb:
http://www.slideshare.net/coolstuff/the-brand-gap

Branding is about consumer trust. If they trust your company to do something consistently well, they will place a good deal of faith in that (hopefully without becoming fanboys). As a result, whenever a product buying decision comes up, they are more likely to select the branded product.

A short introduction to branding

Branding can also work for open source. When people come to trust a "product," or piece of software like FireFox, they will keep using it until given a reason to do otherwise.

The creator of Twitter (as far as I know, one of the most heavily loaded rails apps out there) gave a presentation on scaling at RailsConf this year. The link to the slideshow is below, unfortunately I don't know how much one can get out of it without sound.

However, sites like http://www.twitter.com/ and http://www.revolutionhealth.com/ are dealing quite well with the scaling issues of Rails. Although it may take a bit more ingenuity, many feel that the substantial productivity boost is worth it. Also, as compared to PHP, Rails is definitely a newcomer. Give it a little time.

Scalling Twitter: http://www.slideshare.net/Blaine/scaling-twitter

Was part of panel at SXSW 2007 "Every Breath You Take: Identity, Attention, Reputation and Presence" - Quite an interesting topic.
My presentation here: http://www.slideshare.net/ted.nadeau/sxsw-2007-rep utation-20 ... The internet is still more infantile than infinite.