You have to wonder if they perhaps had a deal with the journalist or the magazine not to publish until after the IPO happened, and then they decided to print anyway.
I have seen this happen in science reporting, unfortunately. A journalist wanted to know about some current work at our department, and got the interview on condition that she wait to publish for a week, until a set of experiments on volunteers had been done (so the volunteers couldn't read the interview and get clued in on what the purpose was). She ignored the deal and ran it just a couple of days later ("we really needed a piece that day"). The experiments had to be postponed for six months and new volunteers had to be found.
Moral: never, ever, tell a journalist about anything with other than historical interest. If any aspect of your work or personal life could be harmed by the timing or manner in which something is published, don't share it. If it is ongoing work, don't speak about it - let your papers do the speaking. Another good, hard-won lesson is: don't make guesses, and don't share your beliefs or estimates unless they are very well covered by your data already. If you feel the need to add "perhaps", or "in my view", or "one possibility is" - just keep quiet. Far too often, that conditional will be dropped once the piece sees print, and your personal opinion will suddenly stand there as scientific fact.
My rant seems to have gone offtopic a bit; feel free to moderate down.
I'm happy Apple never did dominate the market, whatever the reason. Having Microsoft in that role is bad enough, but a secretive, litigation-happy company like Apple with monopoly powers would have been a disaster.
For me it is difficult to put my finger on exactly what has hampered Java's uptake in the general open source community.
It's unlikely to be only one reason. These issues are big, complex with many aspects, and every developer that has made a choice in this field probably have their own unique blend of reasons.
For me, it has come down to a few things, but these hase tended to change as well.
I started playing with Java looong ago, like 1995ish, and actually wrote a small app as part of a summer job (which didn't really go anywhere). It was pretty horrible at the time. A big problem was that we were doing client-side apps, with an UI, and with Java and its UI libs, our then modern machines ended up with the performance of a CBM64, but with far uglier user interface. I still have dreams about that experience after a night with too much beer and rich food.
Today, the performance is better. Using Swing (is it? I mix them up), it tends to look better as well. But: any UI is still uncoupled from the rest of my desktop. I have my nice AA fonts everywhere - except in a Java app, which uses its own font settings and no AA. Controls, cutting and pasting and so on also reinforces that the app is just a free-floating guest on my machine and is not integrated one bit. Also, the runtime takes a _lot_ of resources - on disk and in memory. There sould be no need for that, really - all other VM:s I have (mono, perl, python) seem far less resource hungry.
Oh, and the install is also "too good" for my machine, and plonks down itself in its own private directory, not deigning to play nice with the rest of the machine. If all my other apps can have common resources in/usr/share, libraries in/usr/lib and so on, why can't Java?
OK, this sounds like a litany. It's not that bad, but you wanted to know why people aren't enamoured with Java the way they seem to become about mono, and this is my personal (partial) answer. In short, I write a GTK# app in mono, and it feels like a natural part of my desktop. I write it in Java, and it feels like an intruder.
[...] de Icaza took the interview as an opportunity to lecture managers on why Microsoft should abandon its multibillion-dollar business model and embrace open-source programming. Not surprisingly, de Icaza wasn't hired.
The blurb here makes it sound like he was begging on his knees for them to take him on. Not quite what the article describes. He's not the least "confused on what side he's on".
My big beef with SuSE is that their Gnome packages are somewhat substandard in quality. Don't know exactly what they've done, but it feels pretty clunky and disintegrated compared to Gnome as shipped in Fedora or jhbuild. Here's hoping that it will improve, now that SuSE is part of Novell!
I really like Fedora, but, as you say, it's probably too cutting edge to have a place as a preinstall like this.
1. Jettison the whole gconf/registry thing in favor of a tree of plain text config files in.gnome or something
gconf _is_ a tree of text config files in.gconf.
2. Resurrect the old GNOME control center
What did it do differently than the preferences view in nautilus?
3. Give me a default window manager with the ability to select focus-follows-mouse mouse
Settings->Windows, choose "Focus follows Mouse".
4. Construct a usable menu editor somewhere so that I can customize my menus
Not sure what you feel is wrong with the current method?
8. Give me an "advanced mode" to turn on all kinds of extra GUI configuration bells and whistles like keybindings, autoraise, MIME types, etc.
keybindings - in the preferences already. autoraise windows - you find that in the same preference dialog as focus-follows-mouse above. MIME type editor - already exists, improved for 2.8. For other things, gconf-editor _is_ your advanced mode.
we also have the factor that if all the advanced users hate it, then who's going to develop it?
Fortunately, it turns out a lot of advanced users do not hate it at all. Thus we have several hundred regular contributors to the desktop.
To some extent, designing a desktop for non-technical users parallels designing for the disabled - on the face of it, it seems wasteful and counterproductive, but the benefits really extend to even the momst savvy user/fit and healthy person.
You are pretty happy about the access ramps and extra-wide elevator doors when you are moving heavy furniture, and even the most fit person can appreciate the convenience of a manual jar opener for that one jar of jam that just won't budge. Similarily, it is pretty nice not to have to spend time tinkering with your desktop, and instead focus on getting your work done - whether that work is writing ARM assembly code for robotic arm-joint control or writing invitations for your granddaughers birthday party.
Bikers are the hairy, loud, rude, leather-clad baffoons that ride about on hearing-destroying chrome contraptions.
Hmm. Rather stereotypical, wouldn't you think? I may be hairy and at times leather-clad, but I'm generally considered a genial person and my bike sports neither chrome nor excessive exhaust noise.
Kind of like stating for a fact that bicyclists are gear-obsessed spandex-fetishists with the traffic sense of a drunk octopus. It's not true, of course (well, usually not true. Or at least partly wrong quite often) - but then, neither is your statement about motorcyclists above.
Um, your problem with this scenario is what, exactly? Why does it bother you if other take advantage of the gap? You'll lose only a very little time in any case, and you won't be the one to end up in a serial accident.
I often ride a motorcycle in summer, and if someone starts getting too close behind, I prefer to slow down and go out to the right so the assh^H^H^H^Hbusy person can pass me by and harass someone else. Even on a fourty minute commute, more often than not, I'll see the same car no more than a few hundred meters ahead of me if we happened to go for the same exit. He'd gained literally only seconds on a decent length commute by his aggressive driving.
Ok, off-topic, so please mod down as required. That said, at least in my case, I am working on computer modeling of parts of the brain, as a way to determine the validity of the models that we have of the functionality. Take a model of part of the brain (the amygdala-OFC-Hippocampus system, just to take something completely out of the blue), implement, and run simulations designed to be functionally similar to behavioral or nurological experiments done on real animals. If the model can replicate the results, good. If not (the normal case), we need better models, and the way it fails can tell us something about what kind of change to look for.
My background is mainly in CS, but during my graduate studies I ended up studying a lot of neurology and neuroanatomy.
Well, OK, I am a computational neuroscientist (or a roboticist, depending on how you see it), so I'm not utterly lost, but it's certainly not close to my field either. Better to be clear and upfront about things like that.
IANABC. The "molecules" or parts (a protein is all one molecule, really) don't move around of course; they sit in the same sequence in the chain as always. But, as you say, since a protein tends to be a pretty complex thing, there is usually more than one minimum in the energy landscape for it. Our eyes depend on this, for instance: there's a protein in our vision cells that will jump into another (somewhat unstable) minimum when prodded with energy (=light). In its new shape, it will tend to catalyst a reaction that it otherwise doesn't, and the product of this reaction in turn triggers a nerve signal. This local minimum is not very stable though (it is "shallow"), so after a short while, the protein will revert back to its normal shape, ready to react again when light strikes.
In the case of prions, it seems they can act as templates for each other. As they bump into each other, they will tend to act as a mold, effectively lowering the barrier between the two states. The new state is "narrower" but "deeper", so it is easier for one of the normal prions to slip over to the rouge state when molded than the other way around.
In most of the world, phones _are_ sold to consumers. While European service providers also use cheap/free phones to lure customers, there is no obstacle here to use whatever phone you want with a given subscription. Lots of phone stores and home electronic stores have display cases filled with phones of all kinds.
The problem really is created by the manufacturers as much as the providers. The phone has become a fashion item; for quite a lot of people, the phone you use tells others about who you are. Thus people tend to want to get a new phone very often, as fashions and designs change. That drives down prices a lot, as people can't afford to get a new, really expensive, phone every year, and on the other hand, the manufacturers dump the prices of their new models in order to make them the next must-have.
In a sense, it's the SIM card that defines their phone for people - that's the thing that holds their subscription, as well as address lists, phone numbers and so on. The phone hardware it currently sits in is just another fancy shell, to be discarded whenever the next model comes along.
The screen is no better. It's exactly the same. Take it out into the sunshine and you can't see it anymore.
And even without the screen, a tablet is exactly what I don't want. It is too big, the batteries don't last nearly long enoug - and it is a _computer_. I don't want to babysit another desktop just for a text-reader.
You want a palm. A palm with a better daylight screen. (The Zire 71 has some issues in bright light, but I suspect that's due to the screen cover I have on it.)
Nope. Not good enough. I used to use a Palm III. At the moment, I use a Yopy, which is decent as a book reader; the screen is really good for being LCD, and the software is good enough (Dillo is not much of an online browser, but it is really good as a document reader). But try taking any LCD-based display to the beach, for instance. And the format is a bit too small for my taste. The screen should be about postcard-sized, or perhaps even slightly larger (there's the tradeoff with where you can tuck away the device when not in use).
I've looked at the Sony ebook reader with the new paper-like screen technology, though, and the difference in clarity is striking, especially in bright light. It's not that the contrast is any better (it is quite a bit worse, I think), but that it really does feel like reading on paper, not on a screen.
Of course, it being a Sony, you can forget any of the other points I made - you can't transfer your own texts to it, which makes it a non-starter as far as I am concerned. Also, overall, it feels slightly "cheap", like the buttons will break soon, and the white plastic casing will easily get scuffed and scraped.
* A reader that is light, inexpensive, with excellent graphics, that can easily be read in the sun.
* The reader must allow me to upload any text, not just from its own selection. This includes raw text files, html files and pdf. If I can't use it for papers, references and public domain/copyright expired works, it's not much good for me.
* The books need to be _mine_, in the same way that dead-tree versions are today. I can keep the copy for as long as I want, I can make backups to my hearts content, and I can sell it on, or give it away if or when I tire of it. No tying it to a particular reader in other words. I would not appreciate having to rebuy my library, just because my reader up and died.
* Neither books nor reader is to require any kind of interaction with the manufacturer or seller in any way, once I purchased it. I on't want to feel tied down, and I don't want to feel like I'm just borrowing the thing, not owning it.
Cory Doctorow (who reasonably knows a thing or two about electronic publishing) has a pretty good piece disassembling the Gizmodo article here: Ebook column that gets it all wrong
...except not really, unless you use heavy, archive-quality acid-free paper, or, preferably, parchment. Go to a used-book store and take a look at some of the older paperbacks they are sure to have. The paper is dark, stiff and brittle, and that is after only 40 years or so.
As for inks, you really want to use "real" ink that is soaked into the substrate; the normal printing inks will fade. Laser printer or copier printing is even worse. It doesn't soak into the paper, and so will have a tendency to simply fall off the surface as it changes over time.
And pointing to the existence of a codex or scroll that is thousands of years old says very little about the safety and reliability unless you compare it to the number of codices and scrolls that didn't make it. If one text in a thousand survives a millennium, on average, then "safe and reliable" isn't really the words I'd use.
You know? I just had a thought. Why doesn't someone complie a database of equivalency for products, and put it online?
And that site will be up how many milliseconds before the first take-down notices arrive?
In Europe it could be feasible to do, as consumer protection laws are quite a bit stronger. But then, such a site would likely focus on European product lines, not American, so you'd still be screwed to some extent.
A high-school friend worked in a similar place, making shampoo, when she was interning as a chemist. At the plant they added "profitone" - which was their internal joke name for water. They would have exactly the same soap base, but then add different batches of coloring, jelling agents, odorants, and, of course, "profitone".
Make the product more "luxurious": more jelling agent - "Ohh, look how rich and creamy it is!". Make a "sport-type" shampoo: more water and an odor that is more acidic and less flowery.
Effectively, gconf-editor works as the "advanced user dialogue". And digging through is becoming easier, as gconf-editor is being revamped right now (search has been added, for example).
Slashdot Mirror
You have to wonder if they perhaps had a deal with the journalist or the magazine not to publish until after the IPO happened, and then they decided to print anyway.
I have seen this happen in science reporting, unfortunately. A journalist wanted to know about some current work at our department, and got the interview on condition that she wait to publish for a week, until a set of experiments on volunteers had been done (so the volunteers couldn't read the interview and get clued in on what the purpose was). She ignored the deal and ran it just a couple of days later ("we really needed a piece that day"). The experiments had to be postponed for six months and new volunteers had to be found.
Moral: never, ever, tell a journalist about anything with other than historical interest. If any aspect of your work or personal life could be harmed by the timing or manner in which something is published, don't share it. If it is ongoing work, don't speak about it - let your papers do the speaking. Another good, hard-won lesson is: don't make guesses, and don't share your beliefs or estimates unless they are very well covered by your data already. If you feel the need to add "perhaps", or "in my view", or "one possibility is" - just keep quiet. Far too often, that conditional will be dropped once the piece sees print, and your personal opinion will suddenly stand there as scientific fact.
My rant seems to have gone offtopic a bit; feel free to moderate down.
No sarcasm intended. I am quite serious.
I'm happy Apple never did dominate the market, whatever the reason. Having Microsoft in that role is bad enough, but a secretive, litigation-happy company like Apple with monopoly powers would have been a disaster.
For me it is difficult to put my finger on exactly what has hampered Java's uptake in the general open source community.
/usr/share, libraries in /usr/lib and so on, why can't Java?
It's unlikely to be only one reason. These issues are big, complex with many aspects, and every developer that has made a choice in this field probably have their own unique blend of reasons.
For me, it has come down to a few things, but these hase tended to change as well.
I started playing with Java looong ago, like 1995ish, and actually wrote a small app as part of a summer job (which didn't really go anywhere). It was pretty horrible at the time. A big problem was that we were doing client-side apps, with an UI, and with Java and its UI libs, our then modern machines ended up with the performance of a CBM64, but with far uglier user interface. I still have dreams about that experience after a night with too much beer and rich food.
Today, the performance is better. Using Swing (is it? I mix them up), it tends to look better as well. But: any UI is still uncoupled from the rest of my desktop. I have my nice AA fonts everywhere - except in a Java app, which uses its own font settings and no AA. Controls, cutting and pasting and so on also reinforces that the app is just a free-floating guest on my machine and is not integrated one bit. Also, the runtime takes a _lot_ of resources - on disk and in memory. There sould be no need for that, really - all other VM:s I have (mono, perl, python) seem far less resource hungry.
Oh, and the install is also "too good" for my machine, and plonks down itself in its own private directory, not deigning to play nice with the rest of the machine. If all my other apps can have common resources in
OK, this sounds like a litany. It's not that bad, but you wanted to know why people aren't enamoured with Java the way they seem to become about mono, and this is my personal (partial) answer. In short, I write a GTK# app in mono, and it feels like a natural part of my desktop. I write it in Java, and it feels like an intruder.
[...] de Icaza took the interview as an opportunity to lecture managers on why Microsoft should abandon its multibillion-dollar business model and embrace open-source programming. Not surprisingly, de Icaza wasn't hired.
The blurb here makes it sound like he was begging on his knees for them to take him on. Not quite what the article describes. He's not the least "confused on what side he's on".
Well, since you're asking - yes, I think so.
Anyway, the one in Lund, Sweden was in bloom last winter; of course I wasn't in town at the time and missed it.
My big beef with SuSE is that their Gnome packages are somewhat substandard in quality. Don't know exactly what they've done, but it feels pretty clunky and disintegrated compared to Gnome as shipped in Fedora or jhbuild. Here's hoping that it will improve, now that SuSE is part of Novell!
I really like Fedora, but, as you say, it's probably too cutting edge to have a place as a preinstall like this.
1. Jettison the whole gconf/registry thing in favor of a tree of plain text config files in .gnome or something
.gconf .
gconf _is_ a tree of text config files in
2. Resurrect the old GNOME control center
What did it do differently than the preferences view in nautilus?
3. Give me a default window manager with the ability to select focus-follows-mouse mouse
Settings->Windows, choose "Focus follows Mouse".
4. Construct a usable menu editor somewhere so that I can customize my menus
Not sure what you feel is wrong with the current method?
8. Give me an "advanced mode" to turn on all kinds of extra GUI configuration bells and whistles like keybindings, autoraise, MIME types, etc.
keybindings - in the preferences already. autoraise windows - you find that in the same preference dialog as focus-follows-mouse above. MIME type editor - already exists, improved for 2.8. For other things, gconf-editor _is_ your advanced mode.
we also have the factor that if all the advanced users hate it, then who's going to develop it?
Fortunately, it turns out a lot of advanced users do not hate it at all. Thus we have several hundred regular contributors to the desktop.
To some extent, designing a desktop for non-technical users parallels designing for the disabled - on the face of it, it seems wasteful and counterproductive, but the benefits really extend to even the momst savvy user/fit and healthy person.
You are pretty happy about the access ramps and extra-wide elevator doors when you are moving heavy furniture, and even the most fit person can appreciate the convenience of a manual jar opener for that one jar of jam that just won't budge. Similarily, it is pretty nice not to have to spend time tinkering with your desktop, and instead focus on getting your work done - whether that work is writing ARM assembly code for robotic arm-joint control or writing invitations for your granddaughers birthday party.
Bikers are the hairy, loud, rude, leather-clad baffoons that ride about on hearing-destroying chrome contraptions.
:)
Hmm. Rather stereotypical, wouldn't you think? I may be hairy and at times leather-clad, but I'm generally considered a genial person and my bike sports neither chrome nor excessive exhaust noise.
Kind of like stating for a fact that bicyclists are gear-obsessed spandex-fetishists with the traffic sense of a drunk octopus. It's not true, of course (well, usually not true. Or at least partly wrong quite often) - but then, neither is your statement about motorcyclists above.
Oh, and at least I know how to spell "buffoon"
Um, your problem with this scenario is what, exactly? Why does it bother you if other take advantage of the gap? You'll lose only a very little time in any case, and you won't be the one to end up in a serial accident.
I often ride a motorcycle in summer, and if someone starts getting too close behind, I prefer to slow down and go out to the right so the assh^H^H^H^Hbusy person can pass me by and harass someone else. Even on a fourty minute commute, more often than not, I'll see the same car no more than a few hundred meters ahead of me if we happened to go for the same exit. He'd gained literally only seconds on a decent length commute by his aggressive driving.
Ok, off-topic, so please mod down as required. That said, at least in my case, I am working on computer modeling of parts of the brain, as a way to determine the validity of the models that we have of the functionality. Take a model of part of the brain (the amygdala-OFC-Hippocampus system, just to take something completely out of the blue), implement, and run simulations designed to be functionally similar to behavioral or nurological experiments done on real animals. If the model can replicate the results, good. If not (the normal case), we need better models, and the way it fails can tell us something about what kind of change to look for.
My background is mainly in CS, but during my graduate studies I ended up studying a lot of neurology and neuroanatomy.
Well, OK, I am a computational neuroscientist (or a roboticist, depending on how you see it), so I'm not utterly lost, but it's certainly not close to my field either. Better to be clear and upfront about things like that.
IANABC. The "molecules" or parts (a protein is all one molecule, really) don't move around of course; they sit in the same sequence in the chain as always. But, as you say, since a protein tends to be a pretty complex thing, there is usually more than one minimum in the energy landscape for it. Our eyes depend on this, for instance: there's a protein in our vision cells that will jump into another (somewhat unstable) minimum when prodded with energy (=light). In its new shape, it will tend to catalyst a reaction that it otherwise doesn't, and the product of this reaction in turn triggers a nerve signal. This local minimum is not very stable though (it is "shallow"), so after a short while, the protein will revert back to its normal shape, ready to react again when light strikes.
In the case of prions, it seems they can act as templates for each other. As they bump into each other, they will tend to act as a mold, effectively lowering the barrier between the two states. The new state is "narrower" but "deeper", so it is easier for one of the normal prions to slip over to the rouge state when molded than the other way around.
In most of the world, phones _are_ sold to consumers. While European service providers also use cheap/free phones to lure customers, there is no obstacle here to use whatever phone you want with a given subscription. Lots of phone stores and home electronic stores have display cases filled with phones of all kinds.
The problem really is created by the manufacturers as much as the providers. The phone has become a fashion item; for quite a lot of people, the phone you use tells others about who you are. Thus people tend to want to get a new phone very often, as fashions and designs change. That drives down prices a lot, as people can't afford to get a new, really expensive, phone every year, and on the other hand, the manufacturers dump the prices of their new models in order to make them the next must-have.
In a sense, it's the SIM card that defines their phone for people - that's the thing that holds their subscription, as well as address lists, phone numbers and so on. The phone hardware it currently sits in is just another fancy shell, to be discarded whenever the next model comes along.
The screen is no better. It's exactly the same. Take it out into the sunshine and you can't see it anymore.
And even without the screen, a tablet is exactly what I don't want. It is too big, the batteries don't last nearly long enoug - and it is a _computer_. I don't want to babysit another desktop just for a text-reader.
You want a palm. A palm with a better daylight screen. (The Zire 71 has some issues in bright light, but I suspect that's due to the screen cover I have on it.)
Nope. Not good enough. I used to use a Palm III. At the moment, I use a Yopy, which is decent as a book reader; the screen is really good for being LCD, and the software is good enough (Dillo is not much of an online browser, but it is really good as a document reader). But try taking any LCD-based display to the beach, for instance. And the format is a bit too small for my taste. The screen should be about postcard-sized, or perhaps even slightly larger (there's the tradeoff with where you can tuck away the device when not in use).
I've looked at the Sony ebook reader with the new paper-like screen technology, though, and the difference in clarity is striking, especially in bright light. It's not that the contrast is any better (it is quite a bit worse, I think), but that it really does feel like reading on paper, not on a screen.
Of course, it being a Sony, you can forget any of the other points I made - you can't transfer your own texts to it, which makes it a non-starter as far as I am concerned. Also, overall, it feels slightly "cheap", like the buttons will break soon, and the white plastic casing will easily get scuffed and scraped.
* A reader that is light, inexpensive, with excellent graphics, that can easily be read in the sun.
* The reader must allow me to upload any text, not just from its own selection. This includes raw text files, html files and pdf. If I can't use it for papers, references and public domain/copyright expired works, it's not much good for me.
* The books need to be _mine_, in the same way that dead-tree versions are today. I can keep the copy for as long as I want, I can make backups to my hearts content, and I can sell it on, or give it away if or when I tire of it. No tying it to a particular reader in other words. I would not appreciate having to rebuy my library, just because my reader up and died.
* Neither books nor reader is to require any kind of interaction with the manufacturer or seller in any way, once I purchased it. I on't want to feel tied down, and I don't want to feel like I'm just borrowing the thing, not owning it.
I'm waiting...
Cory Doctorow (who reasonably knows a thing or two about electronic publishing) has a pretty good piece disassembling the Gizmodo article here: Ebook column that gets it all wrong
...except not really, unless you use heavy, archive-quality acid-free paper, or, preferably, parchment. Go to a used-book store and take a look at some of the older paperbacks they are sure to have. The paper is dark, stiff and brittle, and that is after only 40 years or so.
As for inks, you really want to use "real" ink that is soaked into the substrate; the normal printing inks will fade. Laser printer or copier printing is even worse. It doesn't soak into the paper, and so will have a tendency to simply fall off the surface as it changes over time.
And pointing to the existence of a codex or scroll that is thousands of years old says very little about the safety and reliability unless you compare it to the number of codices and scrolls that didn't make it. If one text in a thousand survives a millennium, on average, then "safe and reliable" isn't really the words I'd use.
You know? I just had a thought. Why doesn't someone complie a database of equivalency for products, and put it online?
And that site will be up how many milliseconds before the first take-down notices arrive?
In Europe it could be feasible to do, as consumer protection laws are quite a bit stronger. But then, such a site would likely focus on European product lines, not American, so you'd still be screwed to some extent.
A high-school friend worked in a similar place, making shampoo, when she was interning as a chemist. At the plant they added "profitone" - which was their internal joke name for water. They would have exactly the same soap base, but then add different batches of coloring, jelling agents, odorants, and, of course, "profitone".
Make the product more "luxurious": more jelling agent - "Ohh, look how rich and creamy it is!". Make a "sport-type" shampoo: more water and an odor that is more acidic and less flowery.
Perhaps helpful:
http://gnomefiles.org/app.php?soft_id=46
Effectively, gconf-editor works as the "advanced user dialogue". And digging through is becoming easier, as gconf-editor is being revamped right now (search has been added, for example).
To protect intellectual property they might have to chip the originals.
:)
Ouch... It did crack me up, though