[Etementaki8]

.

I don’t recall my last nick used on Emule forum, so I’ve found this one still in use. Emule user since the very beginning and it remains the very best and only client for my taste.

Never had any issues, however, lately I’m experiencing rather strange enigma, and I wanted to share my curiosity with you. Perhaps more elaborate research of your topics would grant me some clues, perhaps, this is not very special case at all, but it is certainly news to me.

Since the end of the last year, at least twice a month, exactly the same material that I never wanted, never downloaded as my choice, constantly finds its way to my Incoming Files folder. I’ve made a habit since long ago to always look into details, to verify the size, and make sure what I’m downloading - is exactly what I’ve been looking for.

And this is what is most strange - material is exactly 1.36 GB in size. And for example, material that was verified to be 350 MB, or even smaller, that was verified to be of genuine title, after downloaded, suddenly becomes this fake 1.36 GB material. Is there any rational explanation for this? Perhaps, something that is consistent with experience from other users? If this would be some "world hit" material, I would understand some "promotional" scheme for someone to disguise it.

I know that everyone has downloaded at least once a fake find, but usually when you look into the details/name, at least one will provide you with hint, that is something is wrong. What makes me curios is that what seem to be a genuine download, of entirely different size, random file, once it is finished, reappears as mentioned 1.36 GM material. How can something "find me" so many times, disguised in what seem to be verified as entirely opposite file? And most of all - why? Is there some "trigger" within Emule, a bug perhaps, that uses some weakness? Although, for life of me, I don’t get the point of "diabolical purpose" being completely boring, insignificant material from 2013. one I would probably never even hear about it.

I have no issues with my OS/software or Emule as such, just this enigma, that is happening since end of 2014. Also, using the last version of Emule, I use secure servers, with all recommended settings, also, after search is over, I’m strictly using KAD.

I haven’t been damaged in any way, this is certainly not the world’s greatest problem, but I would really like to understand the nature of why is this happening now on regular basis, and since it is always the same file, how can this be possible, even after I make sure details/name says opposite?

Thank you for your patience and your suggestions

This post has been edited by Etementaki8: 25 June 2015 - 09:50 PM

[inman]

Quote

also, after search is over, I’m strictly using KAD.

I can't provide an answer to your problem, but using Kad to search is not recommended if you want to avoid fake files. This is because a Kad search returns all the different file names from all the users using key words. A global server search will return the 1-2 most popular file names.

[coluche]

I don't know.

To me it sounds as if your PC is either infected by sth. responsible for this behaviour, or maybe some funny Windows-UAC-virtualstorage-rights problem is going on - windows then auto-restoring files you deleted.

Did you try deleting the file from an admin-console ?

-----------------------------

inman, on 26 June 2015 - 01:06 AM, said:

his is because a Kad search returns all the different file names from all the users using key words.

I wish this was so - but instead a Kad-search will display ONLY those filenames that fulfil the search criteria. So, NO insta-warning about the file being sth. completely different.

This post has been edited by coluche: 26 June 2015 - 01:45 PM

[inman]

Quote

I wish this was so - but instead a Kad-search will display ONLY those filenames that fulfil the search criteria. So, NO insta-warning about the file being sth. completely different.

Yeah, that is what I meant, although I didn't word it that well. All file names will not be returned in one Kad search, but all possible file names can be returned with a Kad search. As I said all the file names from all the users will be returned (meaning unique file names will be returned). So if someone has named something different than what the file truly is, it will appear in a Kad search. From my experience, a global server search is much more accurate (given that the popular file names are usually correct.)

This post has been edited by inman: 26 June 2015 - 08:12 PM

[fox88]

To me described behaviour mostly resembles software infection.

inman, on 26 June 2015 - 02:06 AM, said:

using Kad to search is not recommended if you want to avoid fake files. This is because a Kad search returns all the different file names from all the users using key words.

I would disagree with the idea.
Any search can return fakes; while certain files could be found in KAD only.
There are indications how many peers know that particular name, and later name should be verified in file details dialog.
Otherwise, use only verified links.

[inman]

fox88, on 28 June 2015 - 01:34 AM, said:

To me described behaviour mostly resembles software infection.

inman, on 26 June 2015 - 02:06 AM, said:

using Kad to search is not recommended if you want to avoid fake files. This is because a Kad search returns all the different file names from all the users using key words.

I would disagree with the idea.
Any search can return fakes; while certain files could be found in KAD only.
There are indications how many peers know that particular name, and later name should be verified in file details dialog.
Otherwise, use only verified links.

Of course any search can return fake files but Kad usually returns more files, meaning a higher chance of fake files. From my experience, Kad searches also return different file extensions categories (video etc.) to watch I search for, while server searches do not.

Try it yourself. Do a global server search for something generic like "documentary" and the servers return 308 results but Kad 378 results. However, Kad returned files with a lot more sources - for example one with 600+ sources. There is a very good chance the '600+ sources file' is not what it is described as ("Documentary - Energy - Nikola Tesla - The Life And Times Of A Forgotten Genius (International Wholesale Video))" in the Kad search, otherwise it would have appeared in the global server search. Even searching using other parts of the filename (e.g "Nikola Tesla", "Life And Times Of A Forgotten Genius" etc.) does not bring up the same '600+ sources file' on a global servers search.

Low and behold, the Kad file with 600+ sources was the typical Emule file - one with about 100 different file names, with totally different descriptions, and an almost guaranteed fake. And why did it appear? Because Kad searches for unique file names, whilst the servers return the most popular files. This can have its advantages and disadvantages. As you said, certain files could be found in KAD only.

This post has been edited by inman: 27 June 2015 - 05:08 PM

[fox88]

inman, on 27 June 2015 - 07:58 PM, said:

but Kad usually returns more files

Plain truth: there are more files in KAD.

inman, on 27 June 2015 - 07:58 PM, said:

meaning a higher chance of fake files.

If you select files incorrectly.

inman, on 27 June 2015 - 07:58 PM, said:

Try it yourself. Do a global server search for something generic like "documentary" and the servers return 308 results but Kad 378 results. However, Kad returned files with a lot more sources - for example one with 600+ sources.

Let's see.
I got only 260 sources, not 600+; though it does not matter.
Because those numerous sources were reported by 2 nodes only. Trust value is 4.76
Compare to 8(11) - trust value is 97.00

inman, on 27 June 2015 - 07:58 PM, said:

Low and behold

Lo and behold.
Lo came old loke, not from low.

[inman]

i do not know what trust value 97.00 means but there are the stats for the file: Clearly not very trustworthy that it is that documentary:

[fox88]

inman, on 27 June 2015 - 07:58 PM, said:

Because Kad searches for unique file names, whilst the servers return the most popular files.

Both KAD and servers return files with the names corresponding to your search terms.
Since when that most popular files is true?

inman, on 27 June 2015 - 09:08 PM, said:

i do not know what trust value 97.00 means but there are the stats for the file: Clearly not very trustworthy that it is that documentary:

It is one of the things that might be taken into account when selecting file for download.
Did you get that the mentioned trust values refer to different files, not to your particular documentary?

[inman]

Quote

Both KAD and servers return files with the names corresponding to your search terms.
Since when that most popular files is true?

Since there are more fake files shared (or appear on) on a Kad than on the servers. These corresponding search terms with Kad are NOT the most popular file names. You offer a better explanation than me as to why that is, but the fact that the 'global server search' has greater accuracy than the 'kad search' should be obvious to any experienced Emule user.

I did a global search for "android", and then the same search moments later with Kad. Compare and contrast:

Kad search:


Global search:


As you can see I searched for the same thing but got totally different results depending on if it was a Kad or global search. So, if your statement "both KAD and servers return files with the names corresponding to your search terms" is true there must be another variable elsewhere to explain why both produce such different results. On top of that, there is a huge variation on sources. Since only Kad returns files with more sources, you can safely assume that these sources are not accurate with their file names and are not named according to the most popular file names (which are usually correct.)

Quote

It is one of the things that might be taken into account when selecting file for download.
Did you get that the mentioned trust values refer to different files, not to your particular documentary?

No, but I don't see how that helps entirely. And what does the value "97" actually mean? What does it mean compared with the value "1"? Did you think the file was fake or not based on the value "97"? Why not just paste the ED2K link and bring up the stats for the file, prior to download?

[fox88]

inman, on 28 June 2015 - 08:24 PM, said:

there must be another variable elsewhere to explain why both produce such different results.

In case you do not read: there are more files in KAD. Possibly, less anti-P2P activity in ED2K.
And if you remember: information from other sources than your own mule, is not reliable.

inman, on 28 June 2015 - 08:24 PM, said:

No, but I don't see how that helps entirely.

Have you ever seen guides on the subject of avoiding fakes?
I repeat: that value is just one of the indicators. Higher should be better.

inman, on 28 June 2015 - 08:24 PM, said:

Why not just paste the ED2K link and bring up the stats for the file, prior to download?

Does anybode forbid you to do so?

[inman]

Quote

In case you do not read: there are more files in KAD. Possibly, less anti-P2P activity in ED2K.

Even if there are more files in Kad, that does not mean there should be a higher ratio of fake files to correctly name files than the servers.

Quote

And if you remember: information from other sources than your own mule, is not reliable.

So? I don't understand the relevance of this point.

Quote

Have you ever seen guides on the subject of avoiding fakes?
I repeat: that value is just one of the indicators. Higher should be better.

You did not repeat "higher should be better." This still does not explain the range of numbers (97 out of what? 100?) but I will see the guides. 97 seems quite high for an obvious fake.

Quote

Does anybode forbid you to do so?

Of course not, but you used another indicator, so I assumed it was advantageous to do so. If 97 is a good indication that the file is OK, then I will stick to my way.

This post has been edited by inman: 28 June 2015 - 08:22 PM

[fox88]

inman, on 28 June 2015 - 11:20 PM, said:

Even if there are more files in Kad, that does not mean there should be a higher ratio of fake files to correctly name files than the servers.

That would be smart to change subject.
Did we talk about ratio before? Did you measure that ratio?

inman, on 28 June 2015 - 11:20 PM, said:

So? I don't understand the relevance of this point.

That is, you do not understand, that number of sources, file names etc. are reported by unreliable sources.

inman, on 28 June 2015 - 11:20 PM, said:

You did not repeat "higher should be better."

No.
I repeated: that value is just one of the indicators.
Clear enough?

inman, on 28 June 2015 - 11:20 PM, said:

97 seems quite high for an obvious fake.

Have you seen my list of found files?
You even do not get that it was not about your file.

inman, on 28 June 2015 - 11:20 PM, said:

Of course not, but you used another indicator, so I assumed it was advantageous to do so.

The guides suggest to more than one indicator. The rest is up to you.

[inman]

Quote

That would be smart to change subject.
Did we talk about ratio before? Did you measure that ratio?

I paraphrased you should avoid Kad if you want to avoid fake files. I also said there was a higher chance of fake files in Kad (than the servers). This implies a greater ratio of fake files, given the probability of you selecting a fake is higher, if the ratio of fake:legit file is higher. I don't have to say everything, sometimes it is implied.

I don't need to measure the ratio, I base my views on many years of experience. You can do so if you wish, if you want to prove me wrong. Although, you still have not offered a satisfactory explanation as to why Kad searches bring up more sources and more fakes (since you dismissed my reasoning.) You say don't trust anything that is not from my Emule, but when I add the file to my download (the ones with the most sources on Kad), Emule tells me there are many different file names (as I mentioned with the documentary file). This backs up what I have been saying and is some proof, along with the inconsistent names on the stats for the file, popular files on Kad can't be trusted.

Quote

Have you seen my list of found files?
You even do not get that it was not about your file.

Yet you replied to a direct quoted me talking about my file, which is where the confusion comes from. You still have not explained if your files were fake or not, and consequently if 97 was a good indicator for you. Your files where scored at 97 - so? Useless information if you did not verify your files to show that it is an accurate/inaccurate indicator. Be more clear and explain your figures, and what they actually mean before attacking me for not understanding something that wasn't clear at all.

[fox88]

inman, on 29 June 2015 - 01:35 AM, said:

I paraphrased you should avoid Kad if you want to avoid fake files.

When many people use KAD only, guess how smart is this advice.

inman, on 29 June 2015 - 01:35 AM, said:

I don't need to measure the ratio, I base my views on many years of experience.

Ratio - it is about numbers; it should be at least properly estimated.
Otherwise you produce only meaningless noise.
I have a few years of experience as well. So what?

inman, on 29 June 2015 - 01:35 AM, said:

You can do so if you wish, if you want to prove me wrong.

Wrong. Your idea - you must prove it. Otherwise it is assumed to be false.

inman, on 29 June 2015 - 01:35 AM, said:

Although, you still have not offered a satisfactory explanation as to why Kad searches bring up more sources and more fakes (since you dismissed my reasoning.)

With all your years of experience, how could you miss spamming and other bad servers?

inman, on 29 June 2015 - 01:35 AM, said:

You say don't trust anything that is not from my Emule, but when I add the file to my download (the ones with the most sources on Kad), Emule tells me there are many different file names (as I mentioned with the documentary file). This backs up what I have been saying and is some proof, along with the inconsistent names on the stats for the file, popular files on Kad can't be trusted.

I noticed you are at odds with numbers; now you are doing the same about logic.
We cannot trust anything sent to us; but it does not mean that anything sent to us is incorrect.

inman, on 29 June 2015 - 01:35 AM, said:

Be more clear and explain your figures, and what they actually mean before attacking me for not understanding something that wasn't clear at all.

Is it still unclear to you that one number alone does not make the file good?
The value is calculated based on network data using a bit of common sense.
Higher is better; anything close to 1 is suspicious.

[Etementaki8]

First of all, Inman and Coluche thank you for your response, and please forgive my late reply:

inman, on 26 June 2015 - 12:06 AM, said:

I can't provide an answer to your problem, but using Kad to search is not recommended if you want to avoid fake files. This is because a Kad search returns all the different file names from all the users using key words. A global server search will return the 1-2 most popular file names.

To be more precise, I mostly use Global search, but once I’m finished with searching, I disconnect with them, and only stay connected with KAD. I Hope this clarifies my previous comment.

I’m using KAD search only those rare times, if something I’m looking for is not available with Global search.

coluche, on 26 June 2015 - 02:41 PM, said:

NTo me it sounds as if your PC is either infected by sth. responsible for this behaviour, or maybe some funny Windows-UAC-virtualstorage-rights problem is going on - windows then auto-restoring files you deleted. Did you try deleting the file from an admin-console ?

M0n0wall/AV/OS all verified, no infections whatsoever. Dedicated Emule machine is using XP, while machine from where I’m accessing Emule machine is WS2003. I was surprised how many files was inside Incoming folder, perhaps not reason to blame for mentioned enigma, but I’ve removed those files simply to have a clearer view of what is inside, so if, or when it happens again, I will try to make more sense.

I've tried as "polite" approach in general terms describing "faulty" files. But reading later comments, I was under impression, that showing captures or links is not allowed? So, encouraged by those post, I hope I'm not breaking any rules either.

So this is the file, one I never wanted, one that found my attention, after disguised under different title :

http://img2.picload....ma20.06.201.png
http://img2.picload....ma20.06.201.png
http://picload.org/i...denigma20.06.20

Since result of this in no way damaging, I never wrote or copied those files that were originally searched and then once completed, served again, as "refugee" for this movie.

This post has been edited by Etementaki8: 01 July 2015 - 09:18 PM

[inman]

fox88, on 02 July 2015 - 04:53 AM, said:

inman, on 29 June 2015 - 01:35 AM, said:

I paraphrased you should avoid Kad if you want to avoid fake files.

When many people use KAD only, guess how smart is this advice.

The causation of many using Kad only has nothing to do with the point at hand, rendering your point useless. Many use Kad only because if you download the latest Emule version it doesn't give you any of the current active servers. Many also still use fake servers, which is why a lot of users have "?" under their server names.

Quote

Ratio - it is about numbers; it should be at least properly estimated.
Otherwise you produce only meaningless noise.
I have a few years of experience as well. So what?

So your experience is clearly as outdated as the guides: "Fake files are often spreaded [sic] by companies that want to hurt your P2P experience (like MPAA etc.) or by users who try to spread viruses etc." "Spreaded" - clearly the guides can't even be written in good English, so why trust such outdated garbage?

Did you ever verify the documentary file? I did not either, but my indication of the file stats led me to not download the file. This is based on my experience. All you stated was just another statistic without context, and statistics without context are useless. You stated "anything close to 1 is suspicious" but you really aren't sure until you complete the download. On the other hand I am suspicious enough, based on variation of filenames I posted earlier, to not download the file name and state that is not the documentary. Clearly with your experience, you don't know whether the documentaries you searched for were correct - as you did not state whether you would or would not download the file based on these suspicions. What I am getting at is someone with experience would put your figure "97" into context. Again, with my experience of downloading from Kad, I certainly would not download the documentary file based on my experience.

Finally, you stated you should use more than one indication but only used one yourself which, coming from someone who has a few years of experience, baffles me a little.

Quote

Wrong. Your idea - you must prove it. Otherwise it is assumed to be false.

I really don't care enough to do so. I would also have to download lots of (likely) fake files to prove anything. Even then, you would have to take my word that they are fake. So what is the point of proving anything when it is virtually impossible to do so?
One glance at my screen shots above gives an estimation at how many fake files there are on Kad, since the files are given codecs etc. but are described as RAR files. What is wrong with an estimation in this case?

Quote

With all your years of experience, how could you miss spamming and other bad servers?

Wait, I thought you were arguing AGAINST the fact Kad is worse for fake files? Now you come with with poor reasons as to why Kad is worse for these types of files?

I don't think those reasons are the main cause. The fake files on Kad spread because they are named ~100 different things so they come up in ~100+ different searches. Why do you think, all the fake files on Kad have so many sources? This doesn't explain why a particular search term for a file on Kad, doesn't appear in a global server search. It could be that the user naming the files is sharing it on Kad only, but this still doesn't explain why Kad: A) has more fake files. B ) Returns a few files with far more sources than the servers. Remember how I said fake files spread and gain many sources? This is not a coincidence.

Quote

I noticed you are at odds with numbers; now you are doing the same about logic.
We cannot trust anything sent to us; but it does not mean that anything sent to us is incorrect.

At odds with numbers? Because you can't add context to your figure as any statistician would?
You said "information from other sources than your own mule, is not reliable", so why do we use Peerates stats etc? Of course the stats can be wrong, but this does mean outside sources are not reliable. Thanks for stating the obvious though, I know what trust means.

What good, reliable information can we use that don't come from our own mule? This statement suggests that unless the file extensions come up as incorrect in our Emule, the only way we can verify files is by downloading and checking them ourselves.

Quote

Is it still unclear to you that one number alone does not make the file good?
The value is calculated based on network data using a bit of common sense.
Higher is better; anything close to 1 is suspicious.

Then why only use one number? I thought you were about proving things, instead of being vague. "A bit of common sense" can be applied when you look at ALL the file names and they are ALL different that the file, so all CANNOT possibly be legit files! These are the junk files that you find on Kad!

This post has been edited by inman: 02 July 2015 - 05:59 PM

[fox88]

inman, on 02 July 2015 - 02:57 PM, said:

The causation of many using Kad only has nothing to do with the point at hand, rendering your point useless.

Your point assumed to be false because you could not prove it.

inman, on 02 July 2015 - 02:57 PM, said:

So your experience is clearly as outdated as the guides: "Fake files are often spreaded [sic] by companies that want to hurt your P2P experience (like MPAA etc.) or by users who try to spread viruses etc." "Spreaded" - clearly the guides can't even be written in good English, so why trust such outdated garbage?

Why are you piling up one nonsense point upon another? Like here: mixing up my experience with guides; discussing quality of language.
Unfortunately, technical knowledge, writing skills and desire to write do not always go together.
Why don't you try to make a better guide - if you can?

inman, on 02 July 2015 - 02:57 PM, said:

I really don't care enough to do so.

Then your point is as good as trash.

inman, on 02 July 2015 - 02:57 PM, said:

Wait, I thought you were arguing AGAINST the fact Kad is worse for fake files? Now you come with with poor reasons as to why Kad is worse for these types of files?

Remember, you failed to prove your point. Your point is void.
Therefore, there is no grounds for serious discussion of distinct advantage of one of the networks.

inman, on 02 July 2015 - 02:57 PM, said:

Why do you think, all the fake files on Kad have so many sources?

Have you ever thought that these numbers could be faked?

inman, on 02 July 2015 - 02:57 PM, said:

At odds with numbers? Because you can't add context to your figure as any statistician would?

Context is quite clear. If you are interested in specifics of calculations, please see source code.

inman, on 02 July 2015 - 02:57 PM, said:

You said "information from other sources than your own mule, is not reliable", so why do we use Peerates stats etc?

That is so naive.
Peerates worked for a long time to get reputation. Compare to data sent to you by anonymous source that you never 'seen' before; and possibly will never 'see' again.
But: even data from Peerates should not be treated as the ultimate truth.

inman, on 02 July 2015 - 02:57 PM, said:

Then why only use one number?

Could you keep in mind more than one sentense you just read when trying to understand the meaning of the whole text?
I never said that you must use only one number.

[inman]

Quote

Your point assumed to be false because you could not prove it.

If you think my point is false this still doesn't explain why you counter it with irrelevant points about how Kad is more popular etc. This suggests you think my point holds some validity.

Quote

Why are you piling up one nonsense point upon another? Like here: mixing up my experience with guides; discussing quality of language.
Unfortunately, technical knowledge, writing skills and desire to write do not always go together.
Why don't you try to make a better guide - if you can?

Your experience and guides? You have no experience other than the guides! I do not write a better guide as I do not use official Emule & because not everyone learns through reading crappy, outdated guides. Some of us learn by figuring things out by ourselves trial and error. Is that so hard to comprehend for you, that people learn other than reading someone else's work? You are the one that says not to trust anything other than your own mule, so why trust someone else's dated, poorly worded crap?

Quote

Then your point is as good as trash.

By this logic, the big bang theory must be trash because it is not proven. Theories and 'guestimations' are not trash, especially when proving my point would likely involve me showing screenshots of porn from fake files I downloaded, and comparing them to the file names on Emule.

Quote

Remember, you failed to prove your point. Your point is void.
Therefore, there is no grounds for serious discussion of distinct advantage of one of the networks.

OK, I still found it strange that you changed your point of view on Kad. Initially you said "any search can return fakes" implying there is no distinct advantages, but now you come up with "spamming" as a distinct disadvantage of Kad and the reasoning behind more fakes. With no proof either - I guess this must make your point irrelevant.

Quote

Have you ever thought that these numbers could be faked?

No because the stats of the files on the servers show similar sources. For example, 400+ to my 600+ Kad documentary file, but Kad is more popular, remember?
Plus, as you say "peerates worked for a long time to get reputation."

Quote

Context is quite clear. If you are interested in specifics of calculations, please see source code.

it wasn't even clear what file(s) were described as and what they actually were. It still isn't.

Quote

Could you keep in mind more than one sentense [sic] you just read when trying to understand the meaning of the whole text?
I never said that you must use only one number.

You used only one number and only one method of testing and you are still totally clueless about the how genuine the file is.

This post has been edited by inman: 02 July 2015 - 07:32 PM

[Etementaki8]

And when I thought nothing but clear voyage: It is happening again!
Emule Security server + 800 MB file = 1.36 GB = Jane Austen world of romance, again

There must be a hidden revelation, message of some sort, possibly instrumental for the world as we know it. Surely, someone is shouting: How many times? Will he ever grasp those hidden numbers?

Strategy? For now, lets leave it inside Incoming file, just to possibly confirm, at least some consistency. I'm downloading the same 800 megs file, to confirm if disguise has random, or precise pattern