Think of it like a radio station.. They have a set, rotating schedule(say news, weather, sports) and set items for each topic... If you listen (on a boring, average day) to a local radio station for say, an hour, you will hear redundant stories... They just cycle through their list. It would be the same way with this publishing program, documentaries, news blurbs, whatever, would be set in a rotating schedule, new items are added to the cycle and old items are removed, respectively.
Now for real fun stuff, have redundant channels. Instead of 1 radio station, have 2 stations, offset by 1/2 so that you can get one, if they are on News when you want Sports, flip to the other.
In reality, I want to see a peer-based Usenet for the purpose of publishing articles(either in/. or NPR-quality)... Things like public key encryption to authenticate the authors, tracking, subscribing, commenting, cross-referencing, caching, etc. All from a peer-to-peer program.
The problem with Freenet is that it is famous for not working. Routing in freenet is crippled. We need a fork, a handful of compitent college programmers to sit down and do a complete walk-through of the code, clean things up, and release it. It is just too much of an alpha-mode project for things to really scale and work.
I am an advocate for what Freenet stands for, and will be pushing its use, once it has a stable, scalable, reliable network base.
Speaking of which: Hikaru, develop of FreenetFork, get in touch with me if you can(or know how I can find him)
Re:we need a competitive open source webmail app
on
Gmail Adds Features
·
· Score: 1
How often does this happen?
1. Great freeware app released 2. Community raves, huge demand/popularity 3. All is well 4. Possibility of some level the freeware app to be paid for/PRO version created, etc. 5. Open source community emulates, in a decent, but not as good reimplimentation of the app... The goal is not to be the best, but to emulate the unique feature that made the freeware app so great?
If gmail loses a couple features, or they make a reasonable PRO/paid-for version that has things like pop3 access, no ads, etc... Pay for it, don't emulate it in spite.
All I would want would be a Palm 6 device, with a clean interface and good pre-config, wifi and bluetooth builtin, 1 expansion slot, decent sized flash space builin, and Newton HWR as an addon/replacement for grafitti in the soft keystroke area.
Sub $300, with an emphasis on battery life, no camera and built-in rendezvous connectivity.
rendezvous would be great for near-area wireless net gaming, live near-area IRC, and device discovery/file-sharing on a standard 802.11 network.
Steve Jobs has, FYI, publicly said that Apple to some extent DID put together a PDA, but scrapped the project. Even sony, maker of arguably the best PDAs(atleast for Palms, IMO), is leaving the American PDA market... The money just isn't there.
Not to troll, as the grandparent did, I have an honest question.
Why is out-of-the-box, desktop-ness a low priority for netbsd, while it is so stressed for Linux? Seems like a 'clean' system would more easily be achieved with a 'clean' OS...
I just wish I had an Ubuntu-like netbsd system using those sweet ports system.
I say seal them in a vault, because if Microsoft wanted to patent, say, Windows XP, the source code coudln't exactly go up on the uspto's website, it would be to easy for code to be 'stolen' and put into a competing product. Internet Explorer's code would be suspected of being fed into gecko, and the NT kernel would be suspected of being in Darwin, Linux, Solaris, etc.
It would be too easy for code to get stolen, so it would have to have some sort of abstraction layer, protecting 2 developers from seeing one another's patented (and competing) code. A third-party knowledge observer would have to be used, to compare the two codebases, not the developers themselves.
"Patenting something is the complete opposite: you are disclosing the invention to the world in exchange for a limited monopoly."
I have always wondered, it use to be that an inventor had to disclose blueprints and specifics on how to build the machine he was patenting. If software is to be patented, I think sourcecode should be provided and sealed in a vault (digital and/or physical) until the patent expires, then that sourcecode becomes public domain.
Ofcourse, I also think patents should be scrutinized better, and the standards for prior art and non-obvious should be stressed again, as they once were.
I think it is a secure, and smart move for Linux(and Mac OS X, which I use more often). It has been a long standing mottoe "Don't login as root unless you absolutely need to, and even then, login just for the task at hand at then logout"... Well, why even have a root account then? sudo makes more sense, use sudo and run the command you need, and get rid of the logout - login as root - do task - logout.
You seem to assume that because bittorrent does not do ant routing, privacy assurance, and secreative quadruple handshake wink-and-nod security features, that it somehow is 'less p2p' than Kazaa. You are wrong. Bittorrent is absolutely p2p based.
It is, I grant you, first generation p2p, in that it is centralized at its core(initial discovery,etc), but feeds off peers. That is valid p2p, and it is similar to napster, that was essenitially a giant database built of all the shares, and once you queried the database for a file, it would return it, and you would download directly from that peer. p2p, for sure, but centralized and prone to failure.
Kazaa is second gen p2p. It is decentralized, but not secure.
eMule and similar are varying levels of second gen p2p... distributed/decentralized, but not secure/anonymous.
When the RIAA breaks the walls down on the donkey-based networks, eMule, or other open source projects, will emerge into the third generation and fundamentally crush the hopes the RI/MPAA have. Freenet, Waste, Ants, Mute are 3rd generation p2p projects but are not large enough, tested enough, developed enough, to be reliable and scale to the napster/kazaa size(in the days of yore).
bittorrent is absolutely p2p, how could you possibly think otherwise?
IBM likes linux because it is a standard OS for their mainframe and server systems.. It is free software to sell their expensive hardware. Seems pretty straight forward to me.
Sun? They are IBM wannabes. They want to support Linux so they too can get a free OS to go with their expensive hardware. Problem is, Solaris is technically superior to Linux, so Sun is having trouble replacing a superior inhouse OS, with a will-be-someday-but-isn't-yet-superior outsourced OS. Linux will be there, it will replace the openVMS's, the Solaris's, the HP-UX's of the world.
But it isn't there yet.
Novell? They wants a better, freer, OS for the e-directory and other server software. They don't want to have to support a no longer up-to-date OS. Free outsourcing is fine by them.
Dell and HP are still too reliant on workstation(windows workstations) to be able to leverage and back Linux without getting a fatal blow from Microsoft.
Software Update Service(soon to be WUS, Windows US) is an intranet Windows Update server. Client workstations can point to it for Automatic(scheduled) Updates and installs of patches you provide. SUS doesn't even use WU directly... Our enterprise is blocking v5(and others) of windows update to avoid SP2, but SUS (we are currently testing) pulls down patches without problem.
SUS is only one of many patch management/deployment solutions, and SUS only (only!) deploys Windows 2k/XP/2k3 security patches, critical patches and service packs. Office patches and other Microsoft software, along with custom patches and third-party patches are not supported, and explicitly blocked.
Sun got its reputation of greatness by doing a handful of things really, really well... Among them,
*Existing. This is critical. *Scaling extremely well. Up to 64 and 128 very efficiently, this is one of their best features. *Security/Trusted. Trusted Solaris has government rating saying it can be reliable, secure, and put in seriously hardcore situation where data is crucial to huge organizations.
Upgrade the core internet backbone. Internet not fast enough? Upgrade the routers and fiber(or lay more) running between the internet hubs, ISPs, service providers and core technology hosting companies. The rest is trickle-down effect...
That, and maybe put more caching servers in place.
What you are talking about is a peer-to-peer world wide web. Where a client request a page, www.slashdot.org, and the file(frontend, output of the database-stuff) that is sent to it may or may not come from the actual server hosting/., rather it would come from either the server, or one of the clients that are also browsing the site(or have recently) at that given time..
Smart caching(of the 'end result website). Purgin the cach ever-so-often and doing so efficiently and reliably(not the bogged-down Temporary Internet Files nightmare in Win/IE). Timestamps(to be able to retrieve the most recent version, rsync from somewhat stale version (10 minutes old) to the current version). Extremely efficient algorithms(for distributing the load quickly, so that 'surfing the web' is as fast or faster).
So your data on the DRAID (distributed RAID) is encrypted with your public key, so that only you can decrypy it.
The system should have redudant locations. Similar to the GFS(Google's Filesystem) that has 3 copies of every piece of data(on different computers), for just that reason.
The system should require that you have 1-3 times as much on your system(that is other's data), that you have on other people's computer.
The system should not have a user's data stored on a single computer, rather each file or group of files are on an array of computer, such that a large portion of your data is available, even if a given node is offline.
This would require that for every meg you uploaded, it would be 2-3 megs on the DRAID. For 100 megs of documents, there would be 300 megs worth of copies of those documents, distributed on a dozen+ system. Because of this, you would need to share(host other's files) more than you upload.
Slashdot Mirror
Think of it like a radio station.. They have a set, rotating schedule(say news, weather, sports) and set items for each topic... If you listen (on a boring, average day) to a local radio station for say, an hour, you will hear redundant stories... They just cycle through their list. It would be the same way with this publishing program, documentaries, news blurbs, whatever, would be set in a rotating schedule, new items are added to the cycle and old items are removed, respectively.
Now for real fun stuff, have redundant channels. Instead of 1 radio station, have 2 stations, offset by 1/2 so that you can get one, if they are on News when you want Sports, flip to the other.
You mean like a news co-op?
/. or NPR-quality)... Things like public key encryption to authenticate the authors, tracking, subscribing, commenting, cross-referencing, caching, etc. All from a peer-to-peer program.
In reality, I want to see a peer-based Usenet for the purpose of publishing articles(either in
The problem with Freenet is that it is famous for not working. Routing in freenet is crippled. We need a fork, a handful of compitent college programmers to sit down and do a complete walk-through of the code, clean things up, and release it. It is just too much of an alpha-mode project for things to really scale and work.
I am an advocate for what Freenet stands for, and will be pushing its use, once it has a stable, scalable, reliable network base.
Speaking of which:
Hikaru, develop of FreenetFork, get in touch with me if you can(or know how I can find him)
How often does this happen?
1. Great freeware app released
2. Community raves, huge demand/popularity
3. All is well
4. Possibility of some level the freeware app to be paid for/PRO version created, etc.
5. Open source community emulates, in a decent, but not as good reimplimentation of the app... The goal is not to be the best, but to emulate the unique feature that made the freeware app so great?
If gmail loses a couple features, or they make a reasonable PRO/paid-for version that has things like pop3 access, no ads, etc... Pay for it, don't emulate it in spite.
All I would want would be a Palm 6 device, with a clean interface and good pre-config, wifi and bluetooth builtin, 1 expansion slot, decent sized flash space builin, and Newton HWR as an addon/replacement for grafitti in the soft keystroke area.
Sub $300, with an emphasis on battery life, no camera and built-in rendezvous connectivity.
rendezvous would be great for near-area wireless net gaming, live near-area IRC, and device discovery/file-sharing on a standard 802.11 network.
Steve Jobs has, FYI, publicly said that Apple to some extent DID put together a PDA, but scrapped the project. Even sony, maker of arguably the best PDAs(atleast for Palms, IMO), is leaving the American PDA market... The money just isn't there.
Not to troll, as the grandparent did, I have an honest question.
Why is out-of-the-box, desktop-ness a low priority for netbsd, while it is so stressed for Linux? Seems like a 'clean' system would more easily be achieved with a 'clean' OS...
I just wish I had an Ubuntu-like netbsd system using those sweet ports system.
I say seal them in a vault, because if Microsoft wanted to patent, say, Windows XP, the source code coudln't exactly go up on the uspto's website, it would be to easy for code to be 'stolen' and put into a competing product. Internet Explorer's code would be suspected of being fed into gecko, and the NT kernel would be suspected of being in Darwin, Linux, Solaris, etc.
It would be too easy for code to get stolen, so it would have to have some sort of abstraction layer, protecting 2 developers from seeing one another's patented (and competing) code. A third-party knowledge observer would have to be used, to compare the two codebases, not the developers themselves.
That disregards the part about mathmatical formulas are not patentable... As they are discovered, not created.
Programming is essentially a markup language surrounding mathmatical formulas and thus, should not be patentable.
"Patenting something is the complete opposite: you are disclosing the invention to the world in exchange for a limited monopoly."
I have always wondered, it use to be that an inventor had to disclose blueprints and specifics on how to build the machine he was patenting. If software is to be patented, I think sourcecode should be provided and sealed in a vault (digital and/or physical) until the patent expires, then that sourcecode becomes public domain.
Ofcourse, I also think patents should be scrutinized better, and the standards for prior art and non-obvious should be stressed again, as they once were.
IBM, Apple and Sun don't make software? Wow. News to me.
I think it is a secure, and smart move for Linux(and Mac OS X, which I use more often). It has been a long standing mottoe "Don't login as root unless you absolutely need to, and even then, login just for the task at hand at then logout"... Well, why even have a root account then? sudo makes more sense, use sudo and run the command you need, and get rid of the logout - login as root - do task - logout.
Isn't this also a feature of SELinux?
Not anymore. They used to, but Yahoo has sinced bought a Search Engine company and have used that as the basis of their searches.
how did you measure the change from 65C to 50C? Is there a monitoring tool in MacOS?
You seem to assume that because bittorrent does not do ant routing, privacy assurance, and secreative quadruple handshake wink-and-nod security features, that it somehow is 'less p2p' than Kazaa. You are wrong. Bittorrent is absolutely p2p based.
It is, I grant you, first generation p2p, in that it is centralized at its core(initial discovery,etc), but feeds off peers. That is valid p2p, and it is similar to napster, that was essenitially a giant database built of all the shares, and once you queried the database for a file, it would return it, and you would download directly from that peer. p2p, for sure, but centralized and prone to failure.
Kazaa is second gen p2p. It is decentralized, but not secure.
eMule and similar are varying levels of second gen p2p... distributed/decentralized, but not secure/anonymous.
When the RIAA breaks the walls down on the donkey-based networks, eMule, or other open source projects, will emerge into the third generation and fundamentally crush the hopes the RI/MPAA have. Freenet, Waste, Ants, Mute are 3rd generation p2p projects but are not large enough, tested enough, developed enough, to be reliable and scale to the napster/kazaa size(in the days of yore).
bittorrent is absolutely p2p, how could you possibly think otherwise?
Shakespeare was said to only user about 30,000 words in all of his works... The average human, in his lifetime, uses far fewer.
You want to Google Bomb a dictionary?
IBM likes linux because it is a standard OS for their mainframe and server systems.. It is free software to sell their expensive hardware. Seems pretty straight forward to me.
Sun? They are IBM wannabes. They want to support Linux so they too can get a free OS to go with their expensive hardware. Problem is, Solaris is technically superior to Linux, so Sun is having trouble replacing a superior inhouse OS, with a will-be-someday-but-isn't-yet-superior outsourced OS. Linux will be there, it will replace the openVMS's, the Solaris's, the HP-UX's of the world.
But it isn't there yet.
Novell? They wants a better, freer, OS for the e-directory and other server software. They don't want to have to support a no longer up-to-date OS. Free outsourcing is fine by them.
Dell and HP are still too reliant on workstation(windows workstations) to be able to leverage and back Linux without getting a fatal blow from Microsoft.
Software Update Service(soon to be WUS, Windows US) is an intranet Windows Update server. Client workstations can point to it for Automatic(scheduled) Updates and installs of patches you provide. SUS doesn't even use WU directly... Our enterprise is blocking v5(and others) of windows update to avoid SP2, but SUS (we are currently testing) pulls down patches without problem.
SUS is only one of many patch management/deployment solutions, and SUS only (only!) deploys Windows 2k/XP/2k3 security patches, critical patches and service packs. Office patches and other Microsoft software, along with custom patches and third-party patches are not supported, and explicitly blocked.
A9 uses Google for Web and Image searches... It does add some Amazon links in the web results, but not always.
No, because Google, like Vi, is clearly superior. :)
Sun got its reputation of greatness by doing a handful of things really, really well... Among them,
*Existing. This is critical.
*Scaling extremely well. Up to 64 and 128 very efficiently, this is one of their best features.
*Security/Trusted. Trusted Solaris has government rating saying it can be reliable, secure, and put in seriously hardcore situation where data is crucial to huge organizations.
And you just KNOW Uncle Same will need one of these to search against all that information.
Upgrade the core internet backbone. Internet not fast enough? Upgrade the routers and fiber(or lay more) running between the internet hubs, ISPs, service providers and core technology hosting companies. The rest is trickle-down effect...
That, and maybe put more caching servers in place.
What you are talking about is a peer-to-peer world wide web. Where a client request a page, www.slashdot.org, and the file(frontend, output of the database-stuff) that is sent to it may or may not come from the actual server hosting /., rather it would come from either the server, or one of the clients that are also browsing the site(or have recently) at that given time..
Smart caching(of the 'end result website). Purgin the cach ever-so-often and doing so efficiently and reliably(not the bogged-down Temporary Internet Files nightmare in Win/IE).
Timestamps(to be able to retrieve the most recent version, rsync from somewhat stale version (10 minutes old) to the current version).
Extremely efficient algorithms(for distributing the load quickly, so that 'surfing the web' is as fast or faster).
So your data on the DRAID (distributed RAID) is encrypted with your public key, so that only you can decrypy it.
The system should have redudant locations. Similar to the GFS(Google's Filesystem) that has 3 copies of every piece of data(on different computers), for just that reason.
The system should require that you have 1-3 times as much on your system(that is other's data), that you have on other people's computer.
The system should not have a user's data stored on a single computer, rather each file or group of files are on an array of computer, such that a large portion of your data is available, even if a given node is offline.
This would require that for every meg you uploaded, it would be 2-3 megs on the DRAID. For 100 megs of documents, there would be 300 megs worth of copies of those documents, distributed on a dozen+ system. Because of this, you would need to share(host other's files) more than you upload.