I am not only speaking in theoretical terms, I have experience in these kinds of things. I am sorry you can't follow my logic, but I see no point in explaining that any further / deeper than I already did.

I am not saying it's impossible to design a more efficient method than currently exists, only that the effort you'd have to put into it to implement, is not worth it.

All the "easy" ways of implementing it, will eventually come back at you to bite you in the ass.

edit: What could be more efficient from the site owner's perspective than a large user base already eager to find a reason to report anything? I'd say that is already mighty efficient in terms of early notification and reliability of hit vs miss, at least for phishing links. If one were to prioritize phishing links in the reporting queue, for example, to make the system more efficient, is a different story... (and maybe that is already the case to some extent, I don't know)

Yes, automated systems could potentially be more work for administrators. And with that I rest my case, as I like to think I made my point clear enough regarding why using automated filters is not worth the effort.

Bottom line, everything short of a white list is not going to help a lot.

Regular Expressions can do far far more than just wildcards. But that is not going to help with the problems I mentioned.

And what would be more efficient about that than the current system? I am sure phising links get reported fairly quickly, and more important more reliably than any automatic filter you have.

Automatically hiding is not really worth it, because then you would have all those complaints about why was my post hidden.

It's not about the URL, I can change that at any time to avoid "free" "gift" and such. It's a cat and mouse game. Not worth it.

If you knew anything about programming at all, you would recognize why I spell it "regexp" and not care at all.

I'm sorry, but all those efforts are futile in my opinion.

A blacklist would only help a little more than reporting and blocking the user. Maybe not even that.

Unless your regexp in effect is a blacklist, you will have false positives. And to avoid that you will have to do at the expense of misses.

The changes to the report system would be needed to update a potential blacklist and distinguish false positives from hits from the regexp filter.
Otherwise it wouldn't be any more efficient than the system of reporting and banning today.

It's not the regexp that is the problem in your suggestion. Your system would need a lot of changes to the existing report system to function properly and without more work for admins and mods than they already have.

However clever you are designing your regexp, there will always be false positives and misses. Phishers will very quickly find a way around them. You'd also have to trigger on every URL shortener service there is.

Only feasible way I see would be to use a white list for URLs. But that would restrict linking too much. Who's then to say (and take the time to evaluate) what sites are allowed to link.

Edit: Also, you need to steal less than a bill's hat to cover the costs of setting up a .com domain.

Trade rule #1.5:

Originally posted by TF2Outpost Rules:

1. Your trade should not contain or be purposed for:
... * Irrelevant pictures attached or in notes.

So I would say it is against the rules.

1. The IDs are from Valve / Steam. Each time the item gets altered or traded it gets a new ID. IDs are in consecutive order and increasing with time, thus a higher IDs means a more recent changes to the item / the more recently it was traded.
The original ID is the very first ID the item ever had before it was traded / altered.

2. Yes, ID = Original ID means original ownership without modifications (like paint, name tag..).

3. Not sure, but I would guess stats.tf does not consider duped / unduped at all. There is no way for them to reliably ascertain if an item is duped or not because they have no direct access to the database but only cached versions of the backpacks.
They would have too many false positives for relatively frequently traded items.

5. The WH was used to quickly sell stolen / scammed items. If the account involved got their backpack restored by Steam Support, the warehouse would then have a "duped" item´(as does the original account, of course, as there is no real difference which item was the original).

I thought the Key price was at around $1.20 when this all started, my mistake. Maybe you could ask your Russian or Ukrainian wholesalers why they had to raise their prices. ;-)

After all, they also compete. And as I understand it never had to rely on the supply within the Steam trading economy anyway. But something must have driven their price up as well, because of lower supply I suppose. But it could also be that their price predominantly went up because of higher demand. The higher demand coming from a lot of normal users recycling Keys in the Steam Market instead of supplying new Mann Co. Keys.

I seriously doubt that they use the steam market themselves, but I could be very wrong here of course.

As always, it's a combination of all factors. Determining which has the most effect, is probably impossible for us to tell. Only Valve and their Greek professor may know a little more, I guess.

Did your turnover in terms of number of Keys increase as well?