That Weasel Reader thing looks great -- *if* you're using a PDA.
There are no good open source projects that I know of to let you read ASCII ebooks on a computer screen.
Constraints I would put on such a project:
* Must support antialiased text. If I'm going to be reading masses of text, I'd rather not see jaggies.
* Must support keyboard and mousewheel navigation.
* Must support some form of good resizing to run in fullscreen mode.
* Must support display with a proportional font. This is harder than it sounds, since proportional display is usually done without a hard-wrapped source, and the PG texts are all hard-wrapped.
* The ability to bookmark locations in the text, and zip back to these saved locations.
* The ability to read gzip- or zip-compressed files. ASCII compresses well, and there's no reason to leave ebooks around uncompressed.
* A find feature. It would be nice if this had glark-style features, so you can do context searches and the like. (actually, it might make a lot of sense to just be a frontend to glark).
* (Optional but nice) the ability to feed output into festival or a similar speech synthesis sytem for listening. Open Source speech synth isn't quite to the point where I'd want to use it for ordinary usage (as opposed to use by the disabled), but it's not awful and some folks may like it.
* (Optional but nice) the ability to remember where you stopped reading.
I've looked at a *lot* of approaches to getting a nice, readable book. This hack takes in a text file and seems to spit out a pretty good pdf viewable in full-scree-mode in xpdf:
#!/bin/bash # Converts a text file into a nice, computer-readable PDF # Usage bookize cat "$@"|tr -d "\r"|enscript -B -f Palatino-Roman24 -M Compscreen --word-wrap -p
"$@".ps ps2pdf "$@".ps && rm "$@".ps
And the required ~/.enscriptrc:
# Media definitions: # name width height llx lly urx ury Media: Compscreen 858 644 0 0 858 644
It is, unfortunately, still not perfect. I've tried writing scripts to feed things in to LaTeX (to enjoy the superior kerning of LaTeX), but I've never been that happy with the results. It's easy to have something that's a metasequence in LaTeX isn't escaped.
Okay, trademarks *also* don't have to be registered (at least in the United States). Simple use "in commerce" (which I would guess PG probably counts as from a legal standpoint) produces an unregistered trademark, which may *optionally* be denoted with "TM".
Plus, this place has zero name recognition. It'd be easy for them to just change their name now, before they put lots of money into advertising (that is, unless they were trying to just make money off of confused people). I'd say that their service is potentially legitimate. I've spent a long time trying to figure out ways to intelligently render long ASCII ebooks into PS/PDF format. It is not easy to do, since a lot of formatting information isn't there any more, and since few people want to read an entire book in a nonproportional font.
Good eye. If you are correct, it *may* be infringement of Adobe's trademark as well -- I'm pretty sure you can't make up ads for other people's products without disclosing who you are (could be wrong on this, but I know that political groups always do it).
You can trademark a generic term for use as a brand, as long as it isn't used to refer to a specific other thing already, and you're capitalizing it as Microsoft did with Windows. I can have an "Apple Fruit Orchard". It's usually a poor idea to do this, as it makes the trademark weaker, but there's nothing wrong with it.
It seems that I was also wrong about whether it's possible to file online. You can, in fact, file entirely online. However, apparently this just feeds the stuff to the beaucracy on the other end, and you pay the same price as the paper world and wait the same amount of time -- $335 per trademark per class (plus the renewal fees).
What about plain old textual ones? I can understand the issue when it comes to graphical trademarks.
Also (admittedly, I should research this), but I'm not sure if a trademark being registered is a definite, unshakeable legal claim that cannot later be challenged. It certainly isn't for patents. So it doesn't have to be perfectly infallible. I would think that it's reasonable for a system to give me "garblelublubschopass" as a trademark, since I'm quite sure that there isn't a single reference on the Web and that nobody else has registered the term.
Hell of a weird-ass place to base a server (on an ADSL line on Maui), when the Project Gutenberg 2 guy is registered as being in either North Carolina (billing whois) or Alaska (admin whois). I'll bet they're regretting it in retrospect, given the slashdotting the thing is getting now.
According to a later post, Project Gutenberg *is* a registered trademark -- I still stand by my complaints about the failings of the trademark system, though.
Project Gutenberg is nonpolitical (and given that they can only use out-of-copyright works, not even particularly topical in their choice of additions).
Project Gutenberg doesn't really directly compete against any companies that I know of, and facilitates people obtaining things in the public domain. I think that Project Gutenberg would be an excellent destination for grant money. If I had some way to vote on US grants going to Project Gutenberg, I certainly would do so.
A few rather nasty people tried trademarking the term "Linux" in various countries, which started a bit of a to-do a ways back. (Note that I'll bet that someone is going to forget about paying the renewal fee in the US, and Linus is going to lose the Linux trademark, and some jackass will grab it again, and the whole thing will start again).
The problem is that, in the United States, a trademark must be registered to recieve any protection. The idea of "trademarks" and "profit-generating businesses" is closely tied together. If you start up a project, you need to drop at *least* $400 or so on a trademark application. It's just not financially feasible to do this for every project on Sourceforge.
There is no sense of "de facto" trademarks, where an institution can recieve protection for a lower amount of money. I suspect Project Gutenberg hasn't registered a trademark, and so they can't go after people hijacking their name.
Frankly, I see no reason why trademark processing on a text trademark (like "Project Gutenberg") should cost more than $10 in a modern, computerized system. It should also be automated, and doable over the Web.
You can't do "Microsoft 2" because "Microsoft" is a registered trademark.
It's kind of depressing how difficult the United States makes it to do gratis projects.
It's also going to involve moving beyond the common vision that "open source MUST be free or it's corrupt!" This is hurting the movement. Open source is NOT great because it's free. It's great because it's open. Plain and simple.
While I know that you meant free as in gratis, this statement sounds enough like "libre" (especially with the juxtaposition of "open source") to make me want to put both you and Stallman in a locked room together.:-)
I think that a big part of the value of open source software *is* that it's freely redistributable. There is a huge difference in the usability of software that is has even a *slight* charge associated with it and stuff that can simply be copied around at will. Maybe micropayment systems will change that, but at the moment, I can recommend PuTTY to anyone on Windows that needs a terminal emulator. I don't have to consider their financial status or budget or whether they need it for a thousand systems. There used to be a number of expensive terminal emulator programs out there, but they have cost associated with them, so they're immediately a second choice to something that you can drop in for free.
Someone who wants to try Linux can do so for free. Someone that wants to try out BeOS (and a *lot* of people have said good things about BeOS) have to put $50 or so on the table (actually, I believe the thing was made gratis a while back, but too late for it to really become a major player).
I believe that gratis horizontal-market software is here to say, and that it was only serious finagling that allowed vendors to keep such a lock on the horizontal market software market. Closing a product's source is one way, as it lets you force a user to see ads or have other undesireable features tied in to a desireable product. Keeping secret protocols or file formats let you sell a product (Hotline, Microsoft Word) as well. Open source severely damages the ability of companies to sell non-gratis software based on barriers to competition that they have erected. That's okay -- software is a market that scales extremely efficiently, so if it's feasible to use an image-editing program that's free, there's no huge costs involved that are tough to resolve from an economic standpoint.
I've got news for all you anti-suit types: Marketing isn't trying to BS someone; it's explaining what your product does, who you've designed it for, and what unique qualities make it better than other choices.
*Amen*. I have never been able to figure out why, the more companies deal with large clients, the more they feel that an obscure description is necessary. I've started to form a theory, however. I've noticed that vendors that work with large clients *always* want to get the large clients on the phone, talking to a salesman, so that they can figure out how to maximize the amount of money they're sticking them for. My suspicion is that vendors feel that if their website's product description is unclear enough (if it has "solution" or "enterprise class" in it, I'd be uncomfortable already) potential clients will call their sales department. I have been in the position of doing purchasing recommendations for two companies I've been at. Perhaps I'm just younger or like using the Internet more than the other people there, but I don't take the approach of "get a salesman's phone number and sit through his schtick" that a lot of other people do. If I can't figure out what a product is or what someone's pricing scheme is in ten minutes from their website, it goes right to the bottom of my list. I'll call someone as a last resort. It's just not worth hassling with the huge quantities of bullshit that salesmen throw at you, having to worry about jotting down anything important they say instead of having a nice textual record to look at, etc.
It's really funny to look at product descriptions on Freshmeat -- the descriptions for commercial products are almost universally worse than open-source projects, probably because commercial types are worried about accidentally limiting their product's capabilities too much. Compare two Freshmeat-listed backup systems -- the commercial Arkeia and the gratis/libre Unison. While each system is related to backups, after reading the Arkeia description, I have a large quantity of bullshit couched in nice adjectives in my head. Despite the fact that Arkeia does a much simpler set of things than Unison does, I have less of an idea of what its capabilities are than of Unison's, partly because Unison's developer didn't waste time with flag-waving.
The problem is that they you get nastiness from the company and lawyers down your back.
Remember the story of HardOCP. People can do some criticism, but outright panning tends to get people crabby. How often do you see minimal scores given for any product in a magazine or on a website? The "no stars" products?
If universities wanted to fund development of BitTorrent to encourage it to choose local hosts (if it didn't already) it could tend to download files from university-local hosts, which would save them money on bandwidth.
Yes, BT's approach of writing to many points in the file is going to tend to fragement things.
My take is...so what? Fragmentation is just not a big deal any more, given the speed of hard disks, more efficient filesystems, and the size of RAM caches. I haven't worried about drive fragmentation in years.
As long as your drive can keep up with a video stream, I can't think of anything else that needs to run in real time.
Traditional downloads will suck up as much bandwidth as possible as well.
The only difference is that theoretically, the remote end may be the limiting factor (which is less likely with BitTorrent), and that BitTorrent also uploads (which kills people on highly asymmetric connections like home DSL without a local cap to keep the modem's buffer from filling up).
Companies are not out modifying BitTorrent. They have no reason to favor MIT over GPL.
The reason BitTorrent is a big deal is:
* It doesn't necessarily easily expose you to tons of pirated content. With Kazaa, pirated copies of Blizzard's games are only a search away.
* It doesn't have spyware/adware/whatnot.
* It integrates nicely with websites. You click, program works.
* Because the interface is from a website, which is effectively a trusted source of information, one doesn't have to worry about having someone search for "World of Warcraft Demo" and finding a hacked bogus copy.
If you really want a computer game magazine you should buy Future Publishing's Edge. It's written for and by adults in an adult style and doesn't go for the prurient.
I dunno. I just went to their website, and one of their four cover stories is "E121: Digital Women Inteviews".
Slashdot Mirror
That Weasel Reader thing looks great -- *if* you're using a PDA.
There are no good open source projects that I know of to let you read ASCII ebooks on a computer screen.
Constraints I would put on such a project:
* Must support antialiased text. If I'm going to be reading masses of text, I'd rather not see jaggies.
* Must support keyboard and mousewheel navigation.
* Must support some form of good resizing to run in fullscreen mode.
* Must support display with a proportional font. This is harder than it sounds, since proportional display is usually done without a hard-wrapped source, and the PG texts are all hard-wrapped.
* The ability to bookmark locations in the text, and zip back to these saved locations.
* The ability to read gzip- or zip-compressed files. ASCII compresses well, and there's no reason to leave ebooks around uncompressed.
* A find feature. It would be nice if this had glark-style features, so you can do context searches and the like. (actually, it might make a lot of sense to just be a frontend to glark).
* (Optional but nice) the ability to feed output into festival or a similar speech synthesis sytem for listening. Open Source speech synth isn't quite to the point where I'd want to use it for ordinary usage (as opposed to use by the disabled), but it's not awful and some folks may like it.
* (Optional but nice) the ability to remember where you stopped reading.
I've looked at a *lot* of approaches to getting a nice, readable book. This hack takes in a text file and seems to spit out a pretty good pdf viewable in full-scree-mode in xpdf:
#!/bin/bash
# Converts a text file into a nice, computer-readable PDF
# Usage bookize
cat "$@"|tr -d "\r"|enscript -B -f Palatino-Roman24 -M Compscreen --word-wrap -p
"$@".ps
ps2pdf "$@".ps && rm "$@".ps
And the required ~/.enscriptrc:
# Media definitions:
# name width height llx lly urx ury
Media: Compscreen 858 644 0 0 858 644
It is, unfortunately, still not perfect. I've tried writing scripts to feed things in to LaTeX (to enjoy the superior kerning of LaTeX), but I've never been that happy with the results. It's easy to have something that's a metasequence in LaTeX isn't escaped.
Damn, I'm *really* whiffing today WRT trademarks.
:-)
Okay, trademarks *also* don't have to be registered (at least in the United States). Simple use "in commerce" (which I would guess PG probably counts as from a legal standpoint) produces an unregistered trademark, which may *optionally* be denoted with "TM".
So, I guess things are okay after all.
Plus, this place has zero name recognition. It'd be easy for them to just change their name now, before they put lots of money into advertising (that is, unless they were trying to just make money off of confused people). I'd say that their service is potentially legitimate. I've spent a long time trying to figure out ways to intelligently render long ASCII ebooks into PS/PDF format. It is not easy to do, since a lot of formatting information isn't there any more, and since few people want to read an entire book in a nonproportional font.
Good eye. If you are correct, it *may* be infringement of Adobe's trademark as well -- I'm pretty sure you can't make up ads for other people's products without disclosing who you are (could be wrong on this, but I know that political groups always do it).
You can trademark a generic term for use as a brand, as long as it isn't used to refer to a specific other thing already, and you're capitalizing it as Microsoft did with Windows. I can have an "Apple Fruit Orchard". It's usually a poor idea to do this, as it makes the trademark weaker, but there's nothing wrong with it.
It seems that I was also wrong about whether it's possible to file online. You can, in fact, file entirely online. However, apparently this just feeds the stuff to the beaucracy on the other end, and you pay the same price as the paper world and wait the same amount of time -- $335 per trademark per class (plus the renewal fees).
What about plain old textual ones? I can understand the issue when it comes to graphical trademarks.
Also (admittedly, I should research this), but I'm not sure if a trademark being registered is a definite, unshakeable legal claim that cannot later be challenged. It certainly isn't for patents. So it doesn't have to be perfectly infallible. I would think that it's reasonable for a system to give me "garblelublubschopass" as a trademark, since I'm quite sure that there isn't a single reference on the Web and that nobody else has registered the term.
Their (small) hosting company is apparently Maui Global Communications.
$ host projectgutenberg.info
projectgutenberg.info has address 207.175.209.175
$ whois 207.175.209.175
[Querying whois.arin.net]
[whois.arin.net]
Genuity GNTY-207-175 (NET-207-175-0-0-1)
207.175.0.0 - 207.175.255.255
Maui Global Communications GTE-CUST-MGC (NET-207-175-209-0-1)
207.175.209.0 - 207.175.213.255
Hell of a weird-ass place to base a server (on an ADSL line on Maui), when the Project Gutenberg 2 guy is registered as being in either North Carolina (billing whois) or Alaska (admin whois). I'll bet they're regretting it in retrospect, given the slashdotting the thing is getting now.
Also notable is the fact that these people didn't even try registering www.projectgutenberg2.info -- they got www.projectgutenberg.info.
This is about as blatant an abuse of the name as I can think of.
According to a later post, Project Gutenberg *is* a registered trademark -- I still stand by my complaints about the failings of the trademark system, though.
Project Gutenberg is nonpolitical (and given that they can only use out-of-copyright works, not even particularly topical in their choice of additions).
Project Gutenberg doesn't really directly compete against any companies that I know of, and facilitates people obtaining things in the public domain. I think that Project Gutenberg would be an excellent destination for grant money. If I had some way to vote on US grants going to Project Gutenberg, I certainly would do so.
A few rather nasty people tried trademarking the term "Linux" in various countries, which started a bit of a to-do a ways back. (Note that I'll bet that someone is going to forget about paying the renewal fee in the US, and Linus is going to lose the Linux trademark, and some jackass will grab it again, and the whole thing will start again).
The problem is that, in the United States, a trademark must be registered to recieve any protection. The idea of "trademarks" and "profit-generating businesses" is closely tied together. If you start up a project, you need to drop at *least* $400 or so on a trademark application. It's just not financially feasible to do this for every project on Sourceforge.
There is no sense of "de facto" trademarks, where an institution can recieve protection for a lower amount of money. I suspect Project Gutenberg hasn't registered a trademark, and so they can't go after people hijacking their name.
Frankly, I see no reason why trademark processing on a text trademark (like "Project Gutenberg") should cost more than $10 in a modern, computerized system. It should also be automated, and doable over the Web.
You can't do "Microsoft 2" because "Microsoft" is a registered trademark.
It's kind of depressing how difficult the United States makes it to do gratis projects.
My guess is that it's one of the Linux distros, something with a big more weight than the floppy or CD standalone "firewall" distributions.
It's also going to involve moving beyond the common vision that "open source MUST be free or it's corrupt!" This is hurting the movement. Open source is NOT great because it's free. It's great because it's open. Plain and simple.
:-)
While I know that you meant free as in gratis, this statement sounds enough like "libre" (especially with the juxtaposition of "open source") to make me want to put both you and Stallman in a locked room together.
I think that a big part of the value of open source software *is* that it's freely redistributable. There is a huge difference in the usability of software that is has even a *slight* charge associated with it and stuff that can simply be copied around at will. Maybe micropayment systems will change that, but at the moment, I can recommend PuTTY to anyone on Windows that needs a terminal emulator. I don't have to consider their financial status or budget or whether they need it for a thousand systems. There used to be a number of expensive terminal emulator programs out there, but they have cost associated with them, so they're immediately a second choice to something that you can drop in for free.
Someone who wants to try Linux can do so for free. Someone that wants to try out BeOS (and a *lot* of people have said good things about BeOS) have to put $50 or so on the table (actually, I believe the thing was made gratis a while back, but too late for it to really become a major player).
I believe that gratis horizontal-market software is here to say, and that it was only serious finagling that allowed vendors to keep such a lock on the horizontal market software market. Closing a product's source is one way, as it lets you force a user to see ads or have other undesireable features tied in to a desireable product. Keeping secret protocols or file formats let you sell a product (Hotline, Microsoft Word) as well. Open source severely damages the ability of companies to sell non-gratis software based on barriers to competition that they have erected. That's okay -- software is a market that scales extremely efficiently, so if it's feasible to use an image-editing program that's free, there's no huge costs involved that are tough to resolve from an economic standpoint.
I've got news for all you anti-suit types: Marketing isn't trying to BS someone; it's explaining what your product does, who you've designed it for, and what unique qualities make it better than other choices.
*Amen*. I have never been able to figure out why, the more companies deal with large clients, the more they feel that an obscure description is necessary. I've started to form a theory, however. I've noticed that vendors that work with large clients *always* want to get the large clients on the phone, talking to a salesman, so that they can figure out how to maximize the amount of money they're sticking them for. My suspicion is that vendors feel that if their website's product description is unclear enough (if it has "solution" or "enterprise class" in it, I'd be uncomfortable already) potential clients will call their sales department. I have been in the position of doing purchasing recommendations for two companies I've been at. Perhaps I'm just younger or like using the Internet more than the other people there, but I don't take the approach of "get a salesman's phone number and sit through his schtick" that a lot of other people do. If I can't figure out what a product is or what someone's pricing scheme is in ten minutes from their website, it goes right to the bottom of my list. I'll call someone as a last resort. It's just not worth hassling with the huge quantities of bullshit that salesmen throw at you, having to worry about jotting down anything important they say instead of having a nice textual record to look at, etc.
It's really funny to look at product descriptions on Freshmeat -- the descriptions for commercial products are almost universally worse than open-source projects, probably because commercial types are worried about accidentally limiting their product's capabilities too much. Compare two Freshmeat-listed backup systems -- the commercial Arkeia and the gratis/libre Unison. While each system is related to backups, after reading the Arkeia description, I have a large quantity of bullshit couched in nice adjectives in my head. Despite the fact that Arkeia does a much simpler set of things than Unison does, I have less of an idea of what its capabilities are than of Unison's, partly because Unison's developer didn't waste time with flag-waving.
The problem is that they you get nastiness from the company and lawyers down your back.
Remember the story of HardOCP. People can do some criticism, but outright panning tends to get people crabby. How often do you see minimal scores given for any product in a magazine or on a website? The "no stars" products?
If universities wanted to fund development of BitTorrent to encourage it to choose local hosts (if it didn't already) it could tend to download files from university-local hosts, which would save them money on bandwidth.
Yes, BT's approach of writing to many points in the file is going to tend to fragement things.
My take is...so what? Fragmentation is just not a big deal any more, given the speed of hard disks, more efficient filesystems, and the size of RAM caches. I haven't worried about drive fragmentation in years.
As long as your drive can keep up with a video stream, I can't think of anything else that needs to run in real time.
Traditional downloads will suck up as much bandwidth as possible as well.
The only difference is that theoretically, the remote end may be the limiting factor (which is less likely with BitTorrent), and that BitTorrent also uploads (which kills people on highly asymmetric connections like home DSL without a local cap to keep the modem's buffer from filling up).
This is flamebait, people. Feel free to look at flopsy's posting history if you don't think he posts a solid and rich set of flamebait.
No, I disagree strongly.
Companies are not out modifying BitTorrent. They have no reason to favor MIT over GPL.
The reason BitTorrent is a big deal is:
* It doesn't necessarily easily expose you to tons of pirated content. With Kazaa, pirated copies of Blizzard's games are only a search away.
* It doesn't have spyware/adware/whatnot.
* It integrates nicely with websites. You click, program works.
* Because the interface is from a website, which is effectively a trusted source of information, one doesn't have to worry about having someone search for "World of Warcraft Demo" and finding a hacked bogus copy.
After the Strawberry Shortcake strip, I think we're all curious to see what Gabe and Tycho do with Playboy.
If you really want a computer game magazine you should buy Future Publishing's Edge. It's written for and by adults in an adult style and doesn't go for the prurient.
I dunno. I just went to their website, and one of their four cover stories is "E121: Digital Women Inteviews".
* Natalie Portman, naked and petrified
* Hot grits down the front of one's pants
* First post