Project Kashmir 2 Community Edition Phase 1
- Thread starter kashmir
- Start date
My draft plan of Project Kashmir Community Edition was posted at the original Project Kashmir main thread yesterday.
I have already received several e-mails from this forum members who are interested to join.
This new project starts on 5/10/2004.
If you want to contribute this project, please send e-mail with the following information to kashmir@covad.net .
1) your e-mail address (required)
2) your ImmigrationPortal.Com user name (optional)
3) your case number w/o last 5 digits (optional)
4) Do you want to become a named contributor ? (yes/no : default no)
If you choose "yes", your ImmigrationPortal.Com user name will be on the list of contributors.
To contribute this project, you need Java 2 Runtime Environment 1.4 or later.
I recommend to prepare it before the project starts.
kashmir myself don't provide a technical support.
But if you post at the supporting thread, you may get a response from somebody of the community.
I have already received several e-mails from this forum members who are interested to join.
This new project starts on 5/10/2004.
If you want to contribute this project, please send e-mail with the following information to kashmir@covad.net .
1) your e-mail address (required)
2) your ImmigrationPortal.Com user name (optional)
3) your case number w/o last 5 digits (optional)
4) Do you want to become a named contributor ? (yes/no : default no)
If you choose "yes", your ImmigrationPortal.Com user name will be on the list of contributors.
To contribute this project, you need Java 2 Runtime Environment 1.4 or later.
I recommend to prepare it before the project starts.
kashmir myself don't provide a technical support.
But if you post at the supporting thread, you may get a response from somebody of the community.
desi-bartender
Registered Users (C)
Kashmir,
I am also trying to do something similar. But I am trying to break up the
tasks by months. I was asking for volunteers who can run my program
for each month. I was starting out with February. A output of such a
program is in the attached zip file.
It would be good if we agree on a common file format. Can you post here
the fields required in the report? Can you take a quick look at my
attached zip file and tell me if I am missing certain fields?
My program also needs JDK14_02 and thats all. It is based on HttpUnit.
If we have two programs that can execute and create results they better
output the same result. It will be good to exchange reports between the
two projects.
Can you outline your plan of this Community edition project? Are you going
to do the scans and just mail out the results?
I am also trying to do something similar. But I am trying to break up the
tasks by months. I was asking for volunteers who can run my program
for each month. I was starting out with February. A output of such a
program is in the attached zip file.
It would be good if we agree on a common file format. Can you post here
the fields required in the report? Can you take a quick look at my
attached zip file and tell me if I am missing certain fields?
My program also needs JDK14_02 and thats all. It is based on HttpUnit.
If we have two programs that can execute and create results they better
output the same result. It will be good to exchange reports between the
two projects.
Can you outline your plan of this Community edition project? Are you going
to do the scans and just mail out the results?
140_takes_4ever
Registered Users (C)
desi-bartender,
I like your format for the data. It is pretty compact and easy to read. One thing I would suggest is putting percentages in brackets next to the numbers for each sub heading and a summary at the end of the month. Similar to what Project Kashmir does.
Kashmir, can we have some standards for this? If there are multiple projects running then they should follow similar formats so people don't get confused.
Not so sure we need to duplicate the effort to have two scanners, but we can split the job between 02/03/04. There are thousands of cases and it will require work from too many people to get the data in the first place let alone duplicating it.
I like your format for the data. It is pretty compact and easy to read. One thing I would suggest is putting percentages in brackets next to the numbers for each sub heading and a summary at the end of the month. Similar to what Project Kashmir does.
Kashmir, can we have some standards for this? If there are multiple projects running then they should follow similar formats so people don't get confused.
Not so sure we need to duplicate the effort to have two scanners, but we can split the job between 02/03/04. There are thousands of cases and it will require work from too many people to get the data in the first place let alone duplicating it.
desi-bartender
Registered Users (C)
Kashmir,
I saw your First Community Edition Draft. I see few problems with it.
1. My biggest concern. You are giving people a scan program without
proxies. I must say that anonymous proxies hide the users behind the
proxies and USCIS site won't log their individual ips. But your scan program
will not do that. So people need to know that they will be seen as accessing
USCIS site.
2. You are breaking the cases at 900. That is a very arbit and random
boundary. I suggest you have better boundaries. Like divide the months
in to three parts because I know some months have like around 3000 cases
in them.
3. Send one person three scanner for each month. His duty will be to scan
the full month in something like two days. One can do this by executing
scanner at work and home.
4. I suggest that people post their result on this site. One can then go
ahead and join the results of many months to get a final output.
5. Consider discussing the output format with Project Anaconda which I am
about to mail out to few people so we don't have disparate outputs which
people can't share. This will eliminate confusion.
6.Also consider using my scanner to do the job or give users the option to
use mine. It uses built in lists of proxies and generates CSV files. Does not
scan Approved cases again and does bunch of optimizations. Does not
suffer from hang problems.
7. I also suggest we wait for another week and get the scanners working
perfectly. I am open to testing your scanner and provide you first inputs
if it is simple and if you need to do something to make it simple. Simplicity
is the key. People will not use it if it is not simple.
8. Agree on a output format. I would like to suggest some changes to
output of Kashmir. We need a few more fields which I always end up
computing in my Excel spreadsheet.
Let me know.
I saw your First Community Edition Draft. I see few problems with it.
1. My biggest concern. You are giving people a scan program without
proxies. I must say that anonymous proxies hide the users behind the
proxies and USCIS site won't log their individual ips. But your scan program
will not do that. So people need to know that they will be seen as accessing
USCIS site.
2. You are breaking the cases at 900. That is a very arbit and random
boundary. I suggest you have better boundaries. Like divide the months
in to three parts because I know some months have like around 3000 cases
in them.
3. Send one person three scanner for each month. His duty will be to scan
the full month in something like two days. One can do this by executing
scanner at work and home.
4. I suggest that people post their result on this site. One can then go
ahead and join the results of many months to get a final output.
5. Consider discussing the output format with Project Anaconda which I am
about to mail out to few people so we don't have disparate outputs which
people can't share. This will eliminate confusion.
6.Also consider using my scanner to do the job or give users the option to
use mine. It uses built in lists of proxies and generates CSV files. Does not
scan Approved cases again and does bunch of optimizations. Does not
suffer from hang problems.
7. I also suggest we wait for another week and get the scanners working
perfectly. I am open to testing your scanner and provide you first inputs
if it is simple and if you need to do something to make it simple. Simplicity
is the key. People will not use it if it is not simple.
8. Agree on a output format. I would like to suggest some changes to
output of Kashmir. We need a few more fields which I always end up
computing in my Excel spreadsheet.
Let me know.
140_takes_4ever
Registered Users (C)
1. My biggest concern. You are giving people a scan program without
proxies. I must say that anonymous proxies hide the users behind the
proxies and USCIS site won't log their individual ips. But your scan program
will not do that. So people need to know that they will be seen as accessing
USCIS site.
Valid, not to promote illegal activities, but hate the idea of people being able to track me down.
2. You are breaking the cases at 900. That is a very arbit and random
boundary. I suggest you have better boundaries. Like divide the months
in to three parts because I know some months have like around 3000 cases
in them.
Actually this is a good break point. 1000 is the max cases per person can scan without proxies. Everyone does exactly the same amount of work.
3. Send one person three scanner for each month. His duty will be to scan
the full month in something like two days. One can do this by executing
scanner at work and home.
I think this is where we disagree, it is patronizing and won't garner much interest if you keep such tight control over a community effort. Either allocate work based upon the volunteers and let one person be responsible right from the start for "x" number of cases which he/she scans continuously on a weekly basis or don't. This 'sending people 3 scanners for each month' idea needs to be reworked.
4. I suggest that people post their result on this site. One can then go
ahead and join the results of many months to get a final output.
Collating is always better than hopping around searching for details. It is much easier to send the data to one person who collates it and puts it in a report format. I would have suggest everyone posting it to a single thread, but then who will monitor chronological order etc. It is too much of a hassle.
6.Also consider using my scanner to do the job or give users the option to
use mine. It uses built in lists of proxies and generates CSV files. Does not
scan Approved cases again and does bunch of optimizations. Does not
suffer from hang problems.
I don't think it matters which scanner we should use. Both can be tested simultaniously and the speed/efficiency/easy of use measured by 2-5 people, and decide upon a standard if required.
proxies. I must say that anonymous proxies hide the users behind the
proxies and USCIS site won't log their individual ips. But your scan program
will not do that. So people need to know that they will be seen as accessing
USCIS site.
Valid, not to promote illegal activities, but hate the idea of people being able to track me down.
2. You are breaking the cases at 900. That is a very arbit and random
boundary. I suggest you have better boundaries. Like divide the months
in to three parts because I know some months have like around 3000 cases
in them.
Actually this is a good break point. 1000 is the max cases per person can scan without proxies. Everyone does exactly the same amount of work.
3. Send one person three scanner for each month. His duty will be to scan
the full month in something like two days. One can do this by executing
scanner at work and home.
I think this is where we disagree, it is patronizing and won't garner much interest if you keep such tight control over a community effort. Either allocate work based upon the volunteers and let one person be responsible right from the start for "x" number of cases which he/she scans continuously on a weekly basis or don't. This 'sending people 3 scanners for each month' idea needs to be reworked.
4. I suggest that people post their result on this site. One can then go
ahead and join the results of many months to get a final output.
Collating is always better than hopping around searching for details. It is much easier to send the data to one person who collates it and puts it in a report format. I would have suggest everyone posting it to a single thread, but then who will monitor chronological order etc. It is too much of a hassle.
6.Also consider using my scanner to do the job or give users the option to
use mine. It uses built in lists of proxies and generates CSV files. Does not
scan Approved cases again and does bunch of optimizations. Does not
suffer from hang problems.
I don't think it matters which scanner we should use. Both can be tested simultaniously and the speed/efficiency/easy of use measured by 2-5 people, and decide upon a standard if required.
Limitation was changed !!
It was reduced to ONLY 250.
I found this fact last night when I checked the log file about the status of proxy servers.
I have to chage the number of one-time scan from 900 to 200.
I think it affects desi-bartender's project, too.
The limitation is not around 1,000 any more.140_takes_4ever said:
It was reduced to ONLY 250.
I found this fact last night when I checked the log file about the status of proxy servers.
I have to chage the number of one-time scan from 900 to 200.
I think it affects desi-bartender's project, too.
Last edited by a moderator:
desi-bartender
Registered Users (C)
This makes my point even stronger for using proxies. For a full yearly scan
of 36,000 cases you will need 180 volunteers.
I would be surprised if you even get 18.
The period needs to be a month. People can take ownership of one month
and I bet we can find 12 volunteers more easily instead of 180. Even if
we find 6 we can do half yearly scan where all the action is.
of 36,000 cases you will need 180 volunteers.
I would be surprised if you even get 18.
The period needs to be a month. People can take ownership of one month
and I bet we can find 12 volunteers more easily instead of 180. Even if
we find 6 we can do half yearly scan where all the action is.
140_takes_4ever
Registered Users (C)
Even with Proxy servers it is going to be a pain in the ass to cycle through so many proxies. I wonder how IP spoofing will work? Is it worth investigating?
kashmir said:
Each contributor scans 200 cases 2 or 3 times per week.kashmir said:
This means we need 100 contributors to scan 60,000 cases per week.
200 x 3 x 100 = 60,000
I try to fully scan as the first cycle of this project this weekend, but the full report will be sent only to contributors in order to encourage participation in this project.
desi-bartender
Registered Users (C)
>>Even with Proxy servers it is going to be a pain in the ass to cycle
>>through so many proxies. I wonder how IP spoofing will work?
>>Is it worth investigating?
How is Proxy servers pain in the ass? Dude my program actually crawls
internet to find proxies. It tests proxies for speed and then it uses those
proxies to scan cases. If one proxy maxes out it switches proxy.
This happens AUTOMATICALLY. You don't have to do anything.
You are just..............starting a program.
How can that be pain
in the behind?
IP Spoofing is easy to do. But it will not work. A two line explanation of
IP spoofing follows.
IP spoofing is a blind attack. You assume the IP of any client and send
packets to server. That sounds just like the thing we need. But the
server response packets go to actually the machine with that ip. You
never get it. So yes you will be able to send a query using any ip to
USCIS address but you won't get any response back.
Trust me. If there was something we could do apart from proxy I would have
already done it.
USCIS can further restrict the number of cases to 100. In that case it will
be impossible to make a scan. 100 is a feel good number. In that case
your volunteer number just expands to 360. Trust me. You can crawl the
internet and find 1000 proxies but you cannot easily find 360 volunteers
and co-ordinate that effort.
>>through so many proxies. I wonder how IP spoofing will work?
>>Is it worth investigating?
How is Proxy servers pain in the ass? Dude my program actually crawls
internet to find proxies. It tests proxies for speed and then it uses those
proxies to scan cases. If one proxy maxes out it switches proxy.
This happens AUTOMATICALLY. You don't have to do anything.
You are just..............starting a program.
How can that be painin the behind?
IP Spoofing is easy to do. But it will not work. A two line explanation of
IP spoofing follows.
IP spoofing is a blind attack. You assume the IP of any client and send
packets to server. That sounds just like the thing we need. But the
server response packets go to actually the machine with that ip. You
never get it. So yes you will be able to send a query using any ip to
USCIS address but you won't get any response back.
Trust me. If there was something we could do apart from proxy I would have
already done it.
USCIS can further restrict the number of cases to 100. In that case it will
be impossible to make a scan. 100 is a feel good number. In that case
your volunteer number just expands to 360. Trust me. You can crawl the
internet and find 1000 proxies but you cannot easily find 360 volunteers
and co-ordinate that effort.
TO: Contributors
You will receive two files:
a) kashmir2scan.zip
b) 20040508NNN.in (NNN is 3-digit number)
Please follows the steps:
1) create a directory (folder) for Project Kashmir2 Community Edition.
Unix/Linux:
$ mkdir kashmir2
Windows:
C:\> md kashmir2
2) save kashmir2scan.zip under the above directory (folder)
3) create a subdirectory named "in".
$ cd kashmir2
$ mkdir in
C:\> cd kashmir2
C:\kashmir2\> md in
4) save 20040508NNN.in under "in" direcotry.
5) execute kashmir2scan.zip at "kashmir2" directory.
$ java -jar kashmir2scan.zip
C:\kashmir2> java -jar kashmir2scan.zip
6) It scans 200 cases, create "out" subdirectory, and store scan results at "out" directory.
7) Send 20040508NNN.out file under "out" directory to kashmir@covad.net .
You will receive two files:
a) kashmir2scan.zip
b) 20040508NNN.in (NNN is 3-digit number)
Please follows the steps:
1) create a directory (folder) for Project Kashmir2 Community Edition.
Unix/Linux:
$ mkdir kashmir2
Windows:
C:\> md kashmir2
2) save kashmir2scan.zip under the above directory (folder)
3) create a subdirectory named "in".
$ cd kashmir2
$ mkdir in
C:\> cd kashmir2
C:\kashmir2\> md in
4) save 20040508NNN.in under "in" direcotry.
5) execute kashmir2scan.zip at "kashmir2" directory.
$ java -jar kashmir2scan.zip
C:\kashmir2> java -jar kashmir2scan.zip
6) It scans 200 cases, create "out" subdirectory, and store scan results at "out" directory.
7) Send 20040508NNN.out file under "out" directory to kashmir@covad.net .
Read the following post:c610 said:
http://immigrationportal.com/showpost.php?p=777197&postcount=6
The list to scan includes 200 case numbers randomly selected.