protection for hub owners

imb · Post by **imb** » 2004-08-11 11:13

i've been thinking recentely how pathetic the current permanent ban system is. nicks are easy to change, a great deal of IPs are dynamic. even a moron can get back into a hub if he's been banned. it's evident something else needs to be done on the client side. if you ban a person, you ban them for a reason. you don't want them there ever again. i think some thought needs to be put into giving each user an individual unhackable number possibly. or something similar. without being a coder, i can see how difficult this would be. i can't pretend to know if it's possible, or be able to contribute any code, but i think it's something that truly needs some thought about. yes i know it would be a long time before every user would have an up to date client, as with the hashing function, but considering most clients are built on dc++ nowadays, i think a start needs to be made. i can see this post being practically ignored, that's fair enough, but i'm sure the coders must agree, hub owners shouldn't have to deal with the issue of users constantly returning, which is almost inevitable in most hubs: when they're already dedicating a phenominal bandwidth and time into running a hub. these sorts of things are very detrimental to the dc network.

second request, which i thought would of been mentioned previously, so did a search for, which yielded no results. option to chose whether a filelist takes focus or not upon download. it's incredibly irritating trying to check users when a list always gets put right in front of you.

third request, a scroll down in the manual kick option, to chose different previous kick messages. again makes it easier in my opinion.

Post by **GargoyleMT** » 2004-08-11 12:13

imb wrote:it's evident something else needs to be done on the client side. ... i can see this post being practically ignored, that's fair enough, but i'm sure the coders must agree, hub owners shouldn't have to deal with the issue of users constantly returning, which is almost inevitable in most hubs: when they're already dedicating a phenominal bandwidth and time into running a hub. these sorts of things are very detrimental to the dc network.

You can't offload this problem to the client in any robust form. For instance, DCStealth would just generate a new ID on being banned.

This post might be practically ignored, but there are very similar discussions that've happened on this board in the past year that cover the same topic - trying to make the client do something not in its self interest. In particular, there is a thread about returning a checksum of the EXE to determine if the user is using a hacked DC++ version.

imb wrote:option to chose whether a filelist takes focus or not upon download. it's incredibly irritating trying to check users when a list always gets put right in front of you.

Code: Select all

 -- 0.4032 2004-08-08 --
* Added options to open file lists and pm's in background (thanks sed)

third request, a scroll down in the manual kick option, to chose different previous kick messages. again makes it easier in my opinion.

You can do this more easily with UserCommands - just make "Kick" the first part of the name, and each kick will show in the "Kick" submenu.

imb · Post by **imb** » 2004-08-11 16:13

well it's limitations on the client side , the hubsoft could hardly identify individual users at the moment. DC Stealth is obsolete and detectable via dcdm. this is not an issue of hacked clients, rather persistant bad users using legitimate clients. but i guess it doesn't matter, as one could rather sneakily use their adl search in an op client to deal with such users. i'm interested how is this not in the clients self interest? and again the issue of a hacked client is a different one, but if you insist it has relevance to this thread, can you please direct me to it? would be interested to read it.

Todi · Post by **Todi** » 2004-08-11 16:30

This would only impact regular users, not the "real" fakers that you actually want to keep out, since they could easily bypass it through changing the code. Normal users don't usually try to bypass bans anyway, so it wouldn't be as useful as you seem to think.

Post by **GargoyleMT** » 2004-08-11 22:59

imb wrote:i'm interested how is this not in the clients self interest?

If you ban me, and I want back in, giving you some unique ID that will let you tell that it's still me (even though I'm on a new IP range with a different nick) is directly against my self interest in getting into the hub.

imb wrote:and again the issue of a hacked client is a different one, but if you insist it has relevance to this thread, can you please direct me to it? would be interested to read it.

Because users who wanted to detect bad clients also wanted DC++ to emit an ID, and check it server side. Anyone who wanted to bypass their check would just send the hash of a known good DC++ version, not the hash of the executable they're running. In both situations, the DC++ extension doesn't help.

[NL]Pur · Post by **[NL]Pur** » 2004-08-13 05:01

If you ban me, and I want back in, giving you some unique ID that will let you tell that it's still me (even though I'm on a new IP range with a different nick) is directly against my self interest in getting into the hub.

This unique ID must be stored localy on the clients computer somewhere, so it's fakable too.

Sir Hitman · Post by **Sir Hitman** » 2004-08-14 01:33

There is nothing that is "unhackable" if you release a version like that in a few days it will be hacked. But there is the posibility to ban by MAC adress (the network cards ID). Then the user would have to change netcard to get back.

But the best is ONLY BE IN REG ONLY HUBS. If there is someone you dont like just remove his account.

Todi · Post by **Todi** » 2004-08-14 04:51

You can't get the MAC adress unless the client sends it, so it's just as worthless as sending a unique ID.

ivulfusbar · Post by **ivulfusbar** » 2004-08-14 06:03

Not many hubs support bans by MAC, some do.. but they are not public. ;))

Post by **GargoyleMT** » 2004-08-14 08:51

Sir Hitman wrote:There is nothing that is "unhackable" if you release a version like that in a few days it will be hacked. But there is the posibility to ban by MAC adress (the network cards ID). Then the user would have to change netcard to get back.

MAC addresses are only exposed on end segments of the network connection - unless you have another service that exposes them (NetBIOS will).

So, basically, getting the MAC to ban with is a decidedly difficult task.

Also, tools like SMAC exist: http://www.klcconsulting.net/smac/

imb · Post by **imb** » 2004-08-20 22:00

Some very good points, understand them all. Perhaps this suggestion has been looked upon in the wrong way, which is partly my fault since i used the word 'unhackable'. We've established that we may never reach 100% as it is at the moment, therefore isn't it best to get that percentage as high as possible? As an example: say the current nick/ip ban had a 50% degree of permenance, and for example, a ban by client hash increased the permenance to 85%, how could that not be positive?

You have a point, a lot of users will not try and return to that hub if they're permenantly banned, but if they truly want back in for whatever reason, it doesn't take much brain power to establish you need to change your ip/nick. It is truly important that bans actually work, gives too much room at the moment for users to piss about regarding their share and attitude towards hubs. Then again, as the file does need to be stored on the persons computer locally, i guess it would make it all pointless... but then again, of course a combination of an ip/nick/client code ban would be positive overall.

I think you smart coder types need to put your heads together on the subject, instead of being so dismissive

Post by **GargoyleMT** » 2004-08-24 10:30

imb wrote:I think you smart coder types need to put your heads together on the subject, instead of being so dismissive

Please don't confuse the lack of agreement with your position with being dismissive. I understand what you want to do and why, but I don't think it's worth it.

ivulfusbar · Post by **ivulfusbar** » 2004-08-24 11:01

every time i read the title for this thread i want to change it into, protection from hub owners. There is a reason for not wanting to tell the hub owners everything about yourself. It is not in my interest. Many hub owners abuse their users in different ways, they forbid users to use this or that without any good reason at all. I will not support them in getting better tools for their sometimes abusive actions.

and-btw-i-run-several-hub-so-im-probably-one-of-those-abusive-hub-owners-myself-ly'ers. ;))

cologic · Post by **cologic** » 2004-08-24 11:37

GargoyleMT wrote:
imb wrote:I think you smart coder types need to put your heads together on the subject, instead of being so dismissive
Please don't confuse the lack of agreement with your position with being dismissive.

On the contrary: I'm simply dismissive.

imb · Post by **imb** » 2004-09-05 10:52

i mean dismissive in the fact that this problem wont even be addressed or discussed, my idea was just an example. i thought the coders could think of something a little more innovative, considering what some of them have done for dc++ already.

ivulfusbar wrote:every time i read the title for this thread i want to change it into, protection from hub owners. There is a reason for not wanting to tell the hub owners everything about yourself. It is not in my interest. Many hub owners abuse their users in different ways, they forbid users to use this or that without any good reason at all. I will not support them in getting better tools for their sometimes abusive actions.

how does this violate anyones privacy, more so than sending your IP out in your client?

? I believe the point in that particular implementation, is to show you are in a public place at your own risk, you don't have the rights you may think you do. are you suggesting bans should not be permenant then? then what is the point in them? how is wanting to keep out those who want to get as much as they can without sharing anything of use, after REPEATED warnings, a hub owner abusing someone?

ivulfusbar wrote:Many hub owners abuse their users in different ways, they forbid users to use this or that without any good reason at all.

i concur, but that point is not relevant at all. if a hub owner or ops abuses it's power in whatever way, then it's simply not a place worth being. this particular idea wont make a difference in that respect.

ivulfusbar wrote:I will not support them in getting better tools for their sometimes abusive actions.

it's run off their pc, and there was me thinking what they say goes (within reason), aslong as they inform what ever it is they may be doing. if oone doesn't agree with it, a user can leave.

GargoyleMT wrote:I understand what you want to do and why, but I don't think it's worth it.

I can understand and appreciate that, and leave it there GargoyleMT but i thought your point, ivulfusbar was absolutely non sensical to me. had to put my piece in.

cologic · Post by **cologic** » 2004-09-05 15:04

Yes, I'm dismissive. I don't view this as a problem to be addressed.

I suspect that indeed "the coders could think of something a little more innovative" than your implementation ideas, but expecting to air a controversial idea without people attacking it on a public forum is a bit silly.

Finally: why should a client act against its own self-interest by helping hubs ban it?

bolamix · Post by **bolamix** » 2004-09-07 06:40

Very interesting topic... As a hubowner and hub-user, I would definitely support a way of making permbans effective. The question of the "client's self interest" is tricky... For me, a hub should be a community of like-minded people who join together to exchange files and chat, within the boundaries set by the hubowners/ops. In this respect, the "client's self interest" is the same as the hub's self interest: to keep leechers and unwanted users/clients out.

... unfinished thinking, dunno where to go from here ... I'll tick "notify me" in case I've inspired anyone

Todi · Post by **Todi** » 2004-09-07 07:04

I think cologic was talking more about why the client would want to provide hub owners with a way to ban it if it was against his self-interest. Regular, good users would probably want that. But you wouldn't want to ban them anyway... someone you want to keep out would probably not want to give you the chance to do so, like already said higher up.

bolamix · Post by **bolamix** » 2004-09-07 11:53

Todi wrote:I think cologic was talking more about why the client would want to provide hub owners with a way to ban it if it was against his self-interest. Regular, good users would probably want that. But you wouldn't want to ban them anyway... someone you want to keep out would probably not want to give you the chance to do so, like already said higher up.

Yup, sorry, it's a never-ending story... But... I don't really care about the opinions of people who don't want to share, so in that respect, I think it'd be better if the ban system (whether hub- or client-side) were more efficient, even if some hackers would undoubtedly find a way around it.
Of course, as a permanent dreamer, i can only wish that everyone was a good user

And I guess I still don't understand why people leech/hack/fake their share

And of course I am no coder and can offer no solution... just tossing ideas around

cologic · Post by **cologic** » 2004-09-07 13:07

bolamix wrote:I don't really care about the opinions of people who don't want to share

I don't really care about the opinions of people who don't want me in their hubs. (Alternatively, this is pointlessly unproductive rhetoric.)

bolamix wrote:so in that respect, I think it'd be better if the ban system (whether hub- or client-side) were more efficient

This doesn't logically follow.

The rest of your post contains only rambling laments, requiring no specific response.

bolamix · Post by **bolamix** » 2004-09-08 00:06

cologic wrote:
bolamix wrote:I don't really care about the opinions of people who don't want to share
I don't really care about the opinions of people who don't want me in their hubs. (Alternatively, this is pointlessly unproductive rhetoric.)

Aw come one now, what is DC all about, if not sharing? And what is the point of coming into a hub if not for sharing? If they don't want to share, why should I operate my hub in their interest? Why should I care about their "own self-interest"?

cologic wrote:
bolamix wrote:so in that respect, I think it'd be better if the ban system (whether hub- or client-side) were more efficient
This doesn't logically follow.

Yes it does. I don't want to operate my hub for the benefit of leechers, so I think being able to ban offenders once for good would be a good thing. More logical now?

Yes, I think a client should be able to "act against its own self-interest", because the interest of the DC community as a whole is more important IMO than the interests of a few leechers.

Anyway as Sir Hitman said, the easiest solution is probably reg-only hubs... it's just that I found the permban discussion pretty interesting, and I didn't understand GargoyleMT's and cologic's logic in why this problem is not worth addressing. I do understand that finding a way to make the ban more effective is a tricky business, and that it cannot be 100% effective. But if it could be made even a little more effective, why not have a go at it, for anyone with enough time and knowledge? (I don't pretend to have either.) Responsible hubowners and users would be thankful, leechers would have a harder time staying in hubs and some would maybe see the benefits of sharing, hackers and fakers would have a new toy to play with... It's in everyone's interests, I'm telling ya

As an aside, I know there is no ready, specific response, but I'm still interested in understanding why some people refuse to share (off-topic, I know, sorry -- is there a "rambling laments" section on this board?)

Todi · Post by **Todi** » 2004-09-08 02:36

Ok... let's use an example here, and see if that helps.

One car dealer sells cars that have a built-in breathalyser that you have to pass before driving, which stops anyone who is drunk to drive them.

One car dealer sells cars without the breathalyser, and people can use that car even when they're drunk.

Obviously, we want everyone to drive the first kind of car, because we don't drive drunk, and we certainly don't want anyone else to. Unfortunately, there is no way to force anyone to drive that kind of a car. Oh we can try to stop them from driving if we see them trying to get into the car, but we can't actually force them to use the kind of car with the breathalyser in it, because it would be against their own self interest, the selfish drunken bastards.

This is the same as with clients. We can make a client that identifies itself so we can ban it, but we can't force fakers to use it unless they volunteer to. And hopefully you now realise how unlikely that is to happen. Fakers want to keep on faking without being banned, and drunk people driving probably won't stop until we throw them in jail or rehabilitation.

DC++ is open source, which means everything we put in can be taken out. Now, if DC was still composed of only NMDC, it would be a different question, since it's closed source. But it probably wouldn't take long for someone to fix that either, since that was how DC++ came to be. It's not that we don't want to stop them, it's just that there is no good way to actually do so. And once you think about it, that's a pretty good thing with the Internet, and we sure as heck don't want that to change.

ivulfusbar · Post by **ivulfusbar** » 2004-09-08 04:08

Todi wrote: DC++ is open source, which means everything we put in can be taken out. Now, if DC was still composed of only NMDC, it would be a different question, since it's closed source. But it probably wouldn't take long for someone to fix that either, since that was how DC++ came to be. It's not that we don't want to stop them, it's just that there is no good way to actually do so. And once you think about it, that's a pretty good thing with the Internet, and we sure as heck don't want that to change.

You are completely right Todi. I can do same kind of cheating with any closed-source client as a open-source by inspecting pacekts and changing them in a proxy.

bolamix · Post by **bolamix** » 2004-09-08 04:40

Todi wrote:Ok... let's use an example here, and see if that helps.

Yup, I dig your example Todi, but I had understood that already. I could probably sum up my position by saying: Knowing that nothing can be perfect, one can still aim for perfection. Some fakers might find it too bothersome to continue faking, just like some regular drunk drivers might change their habits. Note that I said "some", not "most", and that I'm a rather optimistic kinda person (who said "naive"?)

Todi wrote:It's not that we don't want to stop them, it's just that there is no good way to actually do so. And once you think about it, that's a pretty good thing with the Internet, and we sure as heck don't want that to change.

Now that's a much larger debate, and one in which I don't feel very much at ease

Briefly, I feel that a hub is not "the Internet" at large, in that rules on most hubs are pretty specific, as in any public place. If some people don't want to respect them, hubowners have a reasonable right to try and prevent these people from accessing the hub. Even if there is no good way to do that, maybe there's a half-good way, that would be a start, no?

ivulfusbar · Post by **ivulfusbar** » 2004-09-08 04:53

No, life would be much simpler without hubowners. Sure they are usefull for running the software users connects to. But most of the rules / politics they enforce is useless. They think to highly of themself and users needs to be protected from them.

imb · Post by **imb** » 2004-09-08 08:46

Todi wrote:Obviously, we want everyone to drive the first kind of car, because we don't drive drunk, and we certainly don't want anyone else to. Unfortunately, there is no way to force anyone to drive that kind of a car. Oh we can try to stop them from driving if we see them trying to get into the car, but we can't actually force them to use the kind of car with the breathalyser in it, because it would be against their own self interest, the selfish drunken bastards.

Couldn't the same analogy be used with hashing? Still, a lot of people don't want to hash, but eventually these older clients are becoming less and less used. hub owners at the moment could ban non V .400 clients as an example, and the system would work pretty well, would lose about 10% of users in my hub. similary the same would apply with an implemntation to do something to recognise a particular client. after a certain period of time, most would be using the correct client. What other clients are there? Most other clients without faking/bandwidth limiting abilities that you'd want in your hub are built on DC++. including IDC if i'm not mistaken? and that leaves NMDC? which correct me if i'm wrong but the owner has been adding all the dc++ features just to not get his client completely banned from DC. I'm sure he would follow with this feature. Most other clients are out of date or insignificant in my opinion.

There doesn't seem to be anything that can be done with the open source problem.

As i said, ivulfusbar, this feature wont matter in that respect. if hub owners and ops, abuse, they will, with the current kick/ban system. i believe most have common sense and want to make their hub a pleasant place to be with fair rules. anywhere else is not worth being. believe me, i've experienced hubs you're talking about before. they already do abuse and it's not a place i want to stay as the current system is.

what are these useless politics and rules out of interest?

cologic · Post by **cologic** » 2004-09-08 12:32

bolamix wrote:Aw come one now, what is DC all about, if not sharing? And what is the point of coming into a hub if not for sharing? If they don't want to share, why should I operate my hub in their interest? Why should I care about their "own self-interest"?

They must protect their own interests, as you yours.

bolamix wrote:Yes it does. I don't want to operate my hub for the benefit of leechers, so I think being able to ban offenders once for good would be a good thing. More logical now?

Sure.

bolamix wrote:Yes, I think a client should be able to "act against its own self-interest", because the interest of the DC community as a whole is more important IMO than the interests of a few leechers.

Expecting others to uphold your interests merely leads to a less robust system. (Which runs contrary to how I'd define "the interest of the DC community". If you disagree, that merely demonstrates the uselessness of the term.)

To the extend clients/hubs share interests, its in their mutual existence. If hubs start disappearing/becoming unusable, individual client selfishness has backfired. However, if a hub's going to ban a client, that hub's as good as dead anyway to it. (So my ' I don't care about hubs which...' isn't merely reflective/echoing of your statement.)

imb wrote:what are these useless politics and rules out of interest?

The entirety of this:

imb wrote:most would be using the correct client. What other clients are there? Most other clients without faking/bandwidth limiting abilities that you'd want in your hub are built on DC++. including IDC if i'm not mistaken? and that leaves NMDC? which correct me if i'm wrong but the owner has been adding all the dc++ features just to not get his client completely banned from DC. I'm sure he would follow with this feature. Most other clients are out of date or insignificant in my opinion.

<... irrelevant bits snipped ...>

There doesn't seem to be anything that can be done with the open source problem.

But to more directly answer your question about alternative clients: do you dislike all of DC#, DC:PRO, QuickDC, Valknut, and pyDC?

Edit: What open source problem?

bolamix · Post by **bolamix** » 2004-09-08 16:13

cologic wrote:They must protect their own interests, as you yours.

Yes, that I understood. You didn't answer me: what is DC about, if you take the sharing part out of it? I wouldn't mind working in their interest, if their interest was to share.

cologic wrote:Expecting others to uphold your interests merely leads to a less robust system

Unbelievably, yes, I do disagree. Cooperation always leads to better results than competition does, in my and lots of others' experience. All it takes is to realize that there are mutual benefits in sharing, and that by upholding the other party's interests I am upholding mine, too. (That applies to any situation, not just DC.) A kind of double-bind.

cologic wrote:To the extend clients/hubs share interests, its in their mutual existence.

I disagree again. Clients/users and hubs/hubowners have a mutual interest in the quality and circulation of shared files, not just in their mutual existence.

cologic wrote:However, if a hub's going to ban a client, that hub's as good as dead anyway to it. (So my ' I don't care about hubs which...' isn't merely reflective/echoing of your statement.)

Well... here we are talking of the opportunity of preventing banned users from returning to the hubs they were banned from. So it's not that dead, obviously? That said, I'm not sure I understand what you mean in those last two sentences.

As for politics and rules, what I had in mind was users who are banned (after repeated warnings) for:
- sharing unauthorized files
- not sharing at all, incl. preventing others from dlding from them
- faking slots, share, etc

That doesn't seem so useless to my regular users.
I don't know enough about the different clients to start on them, my hub is not on public lists, so I don't get too many fakers...

imb · Post by **imb** » 2004-09-08 17:04

apparentely wanting to keep out clients with bandwidth limiting/tag faking/other faking abilities in a hub with an average dsl connection is the owner abusing it's users. if you didn't get it, a hub isn't like kazaa where 4 million users are logged on and a large percentage aren't even sharing anything, every single user has to be useful in order for the search results and everything else to work well. if you don't want that for you hub, then why else would you run it? wtf? i'm out if that is your point. i give up. i guess i just don't get it.

and individual clients are irrelevant, i was just giving an example. i ban all clients as i said with 'bandwidth limiting/tag faking/other faking abilities'. if anyone else doesn't like that , then it's their problem. it is SUBJECTIVE.

Post by **GargoyleMT** » 2004-09-09 22:37

In some ways, you will get what you want with ADC. In particular, look at the CID.

As the developer of the bandwidth limiting code most clients use, I think I have a different opinion about bad clients than you do. This proposition, and those in the past that intend to enable better bans necessarily involves the topic of banning bad clients. I think, although psf8500 intends for DCDM to be used for good, it's counter-productive. For instance, if I came to your hub, you'd probably ban me right now, though I'm using a development version of DC++. Why would you ban me? Because of the version tag: <++ V:0.40325,M:A,H:0/0/2,S:2>

I hope you see how sucky that is.

cologic · Post by **cologic** » 2004-09-10 10:53

bolamix wrote:what is DC about, if you take the sharing part out of it?

Social interaction? Downloading? Watching hubowners' ludicrous, yet entertaining melodrama? It varies from user to user, and whilst many such attitudes are condemned among hubowners, they exist nonetheless.

bolamix wrote:Cooperation always leads to better results than competition does, in my and lots of others' experience. All it takes is to realize that there are mutual benefits in sharing, and that by upholding the other party's interests I am upholding mine, too.

Yours and others' experience notwithstanding, cooperation often does not benefit involved parties more than competition. (Sharing puts one at risk for copyright enforcement agency targetting, to pick just one example, while downloading effectively doesn't. Competing with the hubowner in his attempts to prevent fakesharing is less legally risky than cooperating with his minimum share and slot requirements.)

bolamix wrote:Clients/users and hubs/hubowners have a mutual interest in the quality and circulation of shared files, not just in their mutual existence.

I'll agree with this, assuming their DC interests stray into downloading. That doesn't require one to want to share one's own files.

bolamix wrote:we are talking of the opportunity of preventing banned users from returning to the hubs they were banned from. So it's not that dead, obviously?

This thread's about increasing bans' effectiveness. Suppose DC had such a system: how would one hub which has banned them and another which has died for all differ? Most hubs I've seen appear to have little unique content, so that client would seldom need to worry about releases filtering out of that hub, even if he himself cannot enter.

bolamix wrote:I don't know enough about the different clients to start on them, my hub is not on public lists, so I don't get too many fakers...

I love how frequently alternative clients get conflated with cheating.

imb wrote:apparentely wanting to keep out clients with bandwidth limiting/tag faking/other faking abilities in a hub with an average dsl connection is the owner abusing it's users.

(1) Average DSL connection? You want what, sympathy? To use that as an excuse to justify arbitrary actions? It's irrelevant.
(2) Who brought up user abuse? I didn't (I have reservations about the phrase). I don't think fusbar did, either.

imb wrote:<... dumbness snipped. DC isn't like KaZaA? I never would have guessed. I view this as the DC analogue of referring to Nazis: it's an easy target, it's wholly irrelevant, and the probability it comes up in discussions such as this approaches 1 ...> every single user has to be useful in order for the search results and everything else to work well. if you don't want that for you hub, then why else would you run it?

I do run a hub, and much as I appreciate the value of people sharing, I do not enjoy dealing with users who must be coerced into doing so. As I involve myself in DC for amusement purposes, this is counterproductive. If the vast majority of users do not of their own volition share, I view that as a failure.

imb wrote:i ban all clients as i said with 'bandwidth limiting/tag faking/other faking abilities'. if anyone else doesn't like that , then it's their problem.

I fully support your right to do this. As a hubowner, you're providing the computer running the hub. You may do anything you wish with it. Just don't expect clients to acquiesce; they're not under your control, cannot be, and should not be.

bolamix · Post by **bolamix** » 2004-09-11 04:04

I'm getting sick of quote-combat, of taking each sentence separately and of arguing for the sake of arguing. Looking at the rest of cologic's posts, he will keep on arguing and taking each sentence separately without reading the whole meaning and putting down his "opponents" with <dumbness snipped> and <irrelevant bits> arguments. He will keep choosing the opposing side just to be on the opposing side. I find this useless so I stop here. Think what you want to think. I've unchecked "Notify me".

GargoyleMT >> I wouldn't ban such a client on sight. I'd ask you in pm what kinda client you're using. As a personal rule I don't ban clients, I ban users using their client in what I view as a dishonest way.

AluminX · Post by **AluminX** » 2004-09-11 04:25

i didn't read allway down but i don't think is hard to ban ppl permanetly from hubs. Make the client to send id hardware to the server and ban em by id hardware instead of ip or username. thats what PunkBuster(PB a game anti-cheat 3rd party program http://www.evenbalance.com) to ban cheatters globally(from all games using PB). cheaters have to change thier hardware to play the game. this is just a thought don't know much about it

Todi · Post by **Todi** » 2004-09-11 04:42

AluminX wrote:i didn't read allway down

You really should have.

Sedulus · Post by **Sedulus** » 2004-09-11 06:36

AluminX:

http://www.shacknews.com/ja.zz?comments=17882 wrote:The whole point is that it updates itself automatically, and up to several timers per day. That's why it can work effectively. Cheats really can't update that fast, unless players trust their cheat to update itself also.

cheats can easily update this fast if the program is open source.

imb · Post by **imb** » 2004-09-11 10:59

i concur with bolamix, instead of taking what is said into context, each sentence is instead dissected on it's own. very juvenile. as an example no i don't expect sympathy... why would i? the MAJORITY of hubs are run on such home connections. where the hub can take about 200-400 users. of course you missed the point completely. that's an average dsl connection. you are only going to find better in few certain areas, such as densely populated areas of sweden etc.

GargoyleMT - i agree with your point, however innocent users will still be kicked/banned occasionally if you were to check manually without the use of such clients. especially as 403 doesn't support the older file lists. those using them need to check for updates all the time, which i do, plus they need to be configured correctly. as to the tag, why not test if i PM you the hub address? will only take 10 secs. i'm curous. could dc++ nnot work a little more closely with i think it's r200 who does the updates for client profiles?

Post by **GargoyleMT** » 2004-09-11 11:48

bolamix wrote:As a personal rule I don't ban clients, I ban users using their client in what I view as a dishonest way.

Good, good. This is sensible. Responsible ops are not the problem, it's irresponsible ops. You can't have a CID-like feature in NMDC protocol hubs for the same reason that DC++ can't have a bandwidth limiting feature - some people will abuse it. (My own opinion, of course.)

imb wrote:GargoyleMT - i agree with your point, however innocent users will still be kicked/banned occasionally if you were to check manually without the use of such clients. especially as 403 doesn't support the older file lists. those using them need to check for updates all the time, which i do, plus they need to be configured correctly.

I do not understand your point here. DC++ will ask users to upgrade to the latest version advertised in Version.xml (0.401) automatically. Users who want to say on the cutting edge can subscribe to the release on Sourceforge's page by clicking on the envelope and get a new email every time a new version of DC++ comes out...

Further, DC++ doesn't typically need any additional configuration from version to version to work properly, only to adjust new features to that release.

It's really the OPs who check users who need to keep up to date the most.

as to the tag, why not test if i PM you the hub address? will only take 10 secs. i'm curous. could dc++ nnot work a little more closely with i think it's r200 who does the updates for client profiles?

Sure I'll do it, but what will that prove? Even if you're not auto-kicking people based on DCDM profiles, some people are. And those are the problem.

PSF8500 does keep up with us, I think, but I haven't heard from him or R200 personally. Since DCDM is their product, and purports to do client identification, it's in their interest to keep their product up to date. If they contact me, I'll do my share (and probably more, but that's just the way I am), but I shouldn't have to be the one to persue them to update it. And even when the client profiles are updated, some OPs still use old DCDM versions, old DC++k versions, and old profiles.

protection for hub owners

protection for hub owners

Re: protection for hub owners