I wanted to let you know about a few legal/money related POPFile things.
1. The POPFile trademark is close to being issued in the US. It has been published for opposition and unless someone actively opposes us I will own the trademark POPFile.
2. Our donation rate has dropped to close to nothing. I estimate that we are receiving about $10 per month at this point. There has been no uptick from the 0.22.3 release; perhaps there will be more from the 0.23.0 when it happens.
3. I am going to keep the money that's been donated towards a Mac and wait until Apple releases their Intel-based Macs. At that point I'll use it towards a new one (and pay the difference myself). I'm hoping that that'll be my primary machine running OS X, Windows Vista and Linux.
4. Junya has been working on an integration with XML-RPC and POPFile that makes POPFile really easy to use in Japan. He's about to release the product and I'll let him describe it in detail in this thread. But from the teams point of view there are a couple of important points: the XML-RPC interface was designed so that the user could integrate without modifying POPFile's code and in a way that would mean that POPFile could be used with non-GPL code. Junya's company are doing that with POPFile. However, I have agreed that in order for them to use the name POPFile in their product they need to pay us a license fee (just for the name, if they modify POPFIle it's still GPL and we get all their changes).
They have agreed to this and will pay me (us) 5% of their revenue on the POPFile-related product.
5. I think that we should move to a slightly more formal way of handling POPFile's money than the current scheme of me getting everything and deciding what to do with it. Does anyone have suggestions of the best way to do that? I'd like to make sure that money we receive from donations, and especially from Junya's company is visible to all POPFile developers and that we decide on the best way to use the money.
John.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I would like to announce that Quick POPFile will be beta released in Japan next week.
Quick POPFile is a windows client that communicates with POPFile and make much easier to reclassify spam and normal messages. It also makes the installation and configuration easy for Outlook Express, Outlook, and Becky, one of the most popular mailer in Japan.
The software is developed so that even a newbie can use it. User does not need to do anything before they can receive the benefit of one of the best open source software.
I joined the POPFile community two years ago. I found that this software was great, but at that time, it could not handle Japanese messages. So, I made a little modification and finally I could make it work with Japanese emails. I was very happy that I could avoid annoying Japanese spam in my mailer, and then I thought that it would be better if other Japanese people could use it. That's why I contributed my modification to POPFile and John and other members in community welcomed this idea. In Japan, now POPFile is known as one of the best spam filter (Yes, I know the power is not limited as spam filter), but it is still for advanced users. I have thought that it will be great if users like my mother or my sister (of course they do not know much about computer) can use POPFile. I joined a new company (a small venture company) this June and the CEO agreed this idea, approved the project to develop a consumer software which makes the installation, configuration and the operation of POPFile much easier.
When I emailed this idea to John, he also agreed, and we made an agreement that we will return 5% of revenue by selling Quick POPFile, to the POPFile community as we use POPFile as a part of the product name (John and the community has the trademark). The official release is planned to be December 1st and the price is planned to be 2500 yen (about $21 in current rate)
On November 7th, it will be beta released and we will offer it to 200 testers by free. The official site is http://www.quickpopfile.jp/\(It cannot be accessed until November 7th, and sorry that it is in Japanese).
I would like to thank to John for understanding this project, and also to the members of the community because without the cooperation of the community, POPFile could not be localized to Japan, and also Quick POPFile could not even exist.
--
Junya ISHIHARA(mailto:junya@unoh.net)
Director
Unoh Inc.(http://www.unoh.net/en_index.html)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
RE 3: Windows Vista? When is Apple going to release those machines? I hope sooner than MS is going to release Vista.
RE 4: This sounds _very_ cool. Without knowing the details, it sounds like a perfect combination of open source and corporate interests.
RE 5: I can't offer any ideas. All that POPFile ever cost me was time. Perhaps all we need to make you feel better is some way to inform the rest of us regularily about the grand total on our imaginary POPFile account?
Manni
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
If we compare the donation rate, the fourm traffic, and major development activity I think we would see a connection. People don't want to pay for software where development looks dead. We know development is continuing in spurts, but regular users don't.
It has been so long since our last release most existing users have already upgraded and have worked out problems. That could explain why it is so quiet in the forums.
It may take a while before people realize PF has been updated. Even this release has so few visual changes people may not see that much has been going on. Hopefully they will read the release notes so they know that bigger development is going on.
I am very interested in Quick POPFile. How does it make things easier? Does it provide built in pre-training?
I don't know how we could better handle the money. We certainly don't need to setup something like the Mozilla foundation. We don't have enough donations or active members to make that work. As long as we get information about what is comming in and how it is used I think leaving it up to John is fine.
Since the global stats aren't working, do we know if we are still gaining users or loosing users? Could that be the problem with fewer donations?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
> Since the global stats aren't working,
> do we know if we are still gaining
> users or loosing users? Could that
> be the problem with fewer donations?
Supposedly, the SF statistics system is bug-ridden, but it is able to produce pretty pictures:
What's striking to me is that we are in the 10,000s of downloads per month and have been for a while. I actually wonder whether the download rate is more related to the number of people who now have some anti-spam solution installed vs 2 years ago when there was a panic to do something.
John.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
> I actually wonder whether the download
> rate is more related to the number of people
> who now have some anti-spam solution
> installed vs 2 years ago when there was
> a panic to do something.
I'm pretty sure that you are right. Not only was there some sort of panic 2 years ago, there was also no commercial competition. The average computer user is still trusting the likes of Microsoft, Symantec, etc. more than he's trusting some OSS project with an ugly home page ;-)
Manni
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
There are a lot more spam solutions out there now. And many of them are integrated into other utility software so it is more convienant. And like Manni says, people trust the big names rather than some open source program.
The other thing is that after running POPFile for a long time people probably are less careful to reclassify mistakes. I know I don't reclassify errors that frequently anymore. I usually do a bunch at a time only every few weeks. If people get tired of reclassifying they may switch to another filter that doesn't require training.
Many ISPs are using better filters now. And lots of people have moved to Gmail or other webbased email since they are getting better.
Even Mozilla Thunderbird has spam filtering built in. It isn't as capable as POPFile, but it is good enough for many people and much simpler to use. Even if our Thunderbird extension ever works, it is not going to be as simple as the built in spam/ham filtering.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm not that sure this shows a loss of users so much as lack of reason to update beyond v0.22.0. There has not been all that much in the last three releases for the average user. If they have no problems you'll not see them downloading.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
>>I am very interested in Quick POPFile. How does it make things easier? Does it provide built in pre-training?
Yes, it provides sample spam emails and classify as spam at the installation. For normal emails, Quick POPFile imports the emails from outbox of user mailer and classify as ham.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
A single user may or may not want to provide funds but a person running a mail server, even a small business, in many cases will pay.
There are still a lot of sites wanting to use POPFile at the mail server level. POPFileD, a daemon for Mercury/32, does a good job of providing the integration but there still is a lot of system admin resistance since there is no real user level correction capability.
POPFile, Mercury/32, POPFileD, and Pegasus Mail work very well together. Pegasus Mail has the capability of using the URL to reclassify but this access is not limited only to a specific users mail.
If there were a real user-by-user network setup I believe you would find a lot more people would be willing to pay for the effort.
If this capability currently exists then it is not apparent to me on how this would be implemented.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Folks,
I wanted to let you know about a few legal/money related POPFile things.
1. The POPFile trademark is close to being issued in the US. It has been published for opposition and unless someone actively opposes us I will own the trademark POPFile.
2. Our donation rate has dropped to close to nothing. I estimate that we are receiving about $10 per month at this point. There has been no uptick from the 0.22.3 release; perhaps there will be more from the 0.23.0 when it happens.
3. I am going to keep the money that's been donated towards a Mac and wait until Apple releases their Intel-based Macs. At that point I'll use it towards a new one (and pay the difference myself). I'm hoping that that'll be my primary machine running OS X, Windows Vista and Linux.
4. Junya has been working on an integration with XML-RPC and POPFile that makes POPFile really easy to use in Japan. He's about to release the product and I'll let him describe it in detail in this thread. But from the teams point of view there are a couple of important points: the XML-RPC interface was designed so that the user could integrate without modifying POPFile's code and in a way that would mean that POPFile could be used with non-GPL code. Junya's company are doing that with POPFile. However, I have agreed that in order for them to use the name POPFile in their product they need to pay us a license fee (just for the name, if they modify POPFIle it's still GPL and we get all their changes).
They have agreed to this and will pay me (us) 5% of their revenue on the POPFile-related product.
5. I think that we should move to a slightly more formal way of handling POPFile's money than the current scheme of me getting everything and deciding what to do with it. Does anyone have suggestions of the best way to do that? I'd like to make sure that money we receive from donations, and especially from Junya's company is visible to all POPFile developers and that we decide on the best way to use the money.
John.
I would like to announce that Quick POPFile will be beta released in Japan next week.
Quick POPFile is a windows client that communicates with POPFile and make much easier to reclassify spam and normal messages. It also makes the installation and configuration easy for Outlook Express, Outlook, and Becky, one of the most popular mailer in Japan.
The software is developed so that even a newbie can use it. User does not need to do anything before they can receive the benefit of one of the best open source software.
I joined the POPFile community two years ago. I found that this software was great, but at that time, it could not handle Japanese messages. So, I made a little modification and finally I could make it work with Japanese emails. I was very happy that I could avoid annoying Japanese spam in my mailer, and then I thought that it would be better if other Japanese people could use it. That's why I contributed my modification to POPFile and John and other members in community welcomed this idea. In Japan, now POPFile is known as one of the best spam filter (Yes, I know the power is not limited as spam filter), but it is still for advanced users. I have thought that it will be great if users like my mother or my sister (of course they do not know much about computer) can use POPFile. I joined a new company (a small venture company) this June and the CEO agreed this idea, approved the project to develop a consumer software which makes the installation, configuration and the operation of POPFile much easier.
When I emailed this idea to John, he also agreed, and we made an agreement that we will return 5% of revenue by selling Quick POPFile, to the POPFile community as we use POPFile as a part of the product name (John and the community has the trademark). The official release is planned to be December 1st and the price is planned to be 2500 yen (about $21 in current rate)
On November 7th, it will be beta released and we will offer it to 200 testers by free. The official site is http://www.quickpopfile.jp/\(It cannot be accessed until November 7th, and sorry that it is in Japanese).
I would like to thank to John for understanding this project, and also to the members of the community because without the cooperation of the community, POPFile could not be localized to Japan, and also Quick POPFile could not even exist.
--
Junya ISHIHARA(mailto:junya@unoh.net)
Director
Unoh Inc.(http://www.unoh.net/en_index.html)
Congratulations, Junya!
This sounds very cool. I had no idea that POPFile was so popular in Japan. I certainly wish it was more popular around here.
Good luck with Quick POPFile.
Manni
Hi John,
RE 1: Good to know.
RE 2: That's not so good. But just how bad is it?
RE 3: Windows Vista? When is Apple going to release those machines? I hope sooner than MS is going to release Vista.
RE 4: This sounds _very_ cool. Without knowing the details, it sounds like a perfect combination of open source and corporate interests.
RE 5: I can't offer any ideas. All that POPFile ever cost me was time. Perhaps all we need to make you feel better is some way to inform the rest of us regularily about the grand total on our imaginary POPFile account?
Manni
If we compare the donation rate, the fourm traffic, and major development activity I think we would see a connection. People don't want to pay for software where development looks dead. We know development is continuing in spurts, but regular users don't.
It has been so long since our last release most existing users have already upgraded and have worked out problems. That could explain why it is so quiet in the forums.
It may take a while before people realize PF has been updated. Even this release has so few visual changes people may not see that much has been going on. Hopefully they will read the release notes so they know that bigger development is going on.
I am very interested in Quick POPFile. How does it make things easier? Does it provide built in pre-training?
I don't know how we could better handle the money. We certainly don't need to setup something like the Mozilla foundation. We don't have enough donations or active members to make that work. As long as we get information about what is comming in and how it is used I think leaving it up to John is fine.
Since the global stats aren't working, do we know if we are still gaining users or loosing users? Could that be the problem with fewer donations?
> Since the global stats aren't working,
> do we know if we are still gaining
> users or loosing users? Could that
> be the problem with fewer donations?
Supposedly, the SF statistics system is bug-ridden, but it is able to produce pretty pictures:
http://sourceforge.net/project/stats/detail.php?group_id=63137&ugn=popfile&type=prdownload&mode=alltime&package_id=0&release_id=0&file_id=0
Judging from those stats, we sure are loosing users.
Manni
What's striking to me is that we are in the 10,000s of downloads per month and have been for a while. I actually wonder whether the download rate is more related to the number of people who now have some anti-spam solution installed vs 2 years ago when there was a panic to do something.
John.
> I actually wonder whether the download
> rate is more related to the number of people
> who now have some anti-spam solution
> installed vs 2 years ago when there was
> a panic to do something.
I'm pretty sure that you are right. Not only was there some sort of panic 2 years ago, there was also no commercial competition. The average computer user is still trusting the likes of Microsoft, Symantec, etc. more than he's trusting some OSS project with an ugly home page ;-)
Manni
There are a lot more spam solutions out there now. And many of them are integrated into other utility software so it is more convienant. And like Manni says, people trust the big names rather than some open source program.
The other thing is that after running POPFile for a long time people probably are less careful to reclassify mistakes. I know I don't reclassify errors that frequently anymore. I usually do a bunch at a time only every few weeks. If people get tired of reclassifying they may switch to another filter that doesn't require training.
Many ISPs are using better filters now. And lots of people have moved to Gmail or other webbased email since they are getting better.
Even Mozilla Thunderbird has spam filtering built in. It isn't as capable as POPFile, but it is good enough for many people and much simpler to use. Even if our Thunderbird extension ever works, it is not going to be as simple as the built in spam/ham filtering.
I'm not that sure this shows a loss of users so much as lack of reason to update beyond v0.22.0. There has not been all that much in the last three releases for the average user. If they have no problems you'll not see them downloading.
>>I am very interested in Quick POPFile. How does it make things easier? Does it provide built in pre-training?
Yes, it provides sample spam emails and classify as spam at the installation. For normal emails, Quick POPFile imports the emails from outbox of user mailer and classify as ham.
So you are focusing only one 2 bucket ham/spam use? Have you been able to improve the reclassification interface with the mail clients?
Just a lurkers view.
A single user may or may not want to provide funds but a person running a mail server, even a small business, in many cases will pay.
There are still a lot of sites wanting to use POPFile at the mail server level. POPFileD, a daemon for Mercury/32, does a good job of providing the integration but there still is a lot of system admin resistance since there is no real user level correction capability.
POPFile, Mercury/32, POPFileD, and Pegasus Mail work very well together. Pegasus Mail has the capability of using the URL to reclassify but this access is not limited only to a specific users mail.
If there were a real user-by-user network setup I believe you would find a lot more people would be willing to pay for the effort.
If this capability currently exists then it is not apparent to me on how this would be implemented.
A real multiuser server based setup is what we are working on for the next major release. That is part of the reason it is taking so long.