Another way to make deep Subtlety decent for PvE would be to duplicate the Relentless Strikes talent on Shadowstep, maybe even add the Ruthlessness effect to Sinister Calling too.
If this was done it would probably need to not stack with the Assasination talents though.
This would mean a 0/2x/4x build wouln't be as handicapped as they currently are.

Adding 10% damage buff to Hemo and Backstab will help an Ass/Sub build in PvP but still doesn't help a PvE one due to duel wield being 20 in combat and Imp Backstab being 8 in Assasination, both are needed to do competative raid damage.

If they are trying to make all of the builds viable in both PvE and PvP swapping Opportunity and Lightning Reflexes would help both Combat Daggers and Mutilate.
Lots of things to consider and I have probably missed something, my rogue is only an alt I don't get to play enough but I find the class quite interesting.

It's easy enough to get Relentless Strikes with a Shadowstep build, but it's marginally useful really. Once you're out of Stealth, your CP generation is essentially less than that of a Combat Dagger rogue (assuming you're going for Backstab/daggers).

Personally, I think Sinister Calling could have Backstab and Hemo add 2 combo points instead of just one. That would give Subt specs a lot more viability in both environments. It'd be easier to keep SnD running in a raid, it's be much easier to lock down an opponent in PVP.

And although it makes sense, if they were to swap Opportunity and Lightning Reflexes, I think I'd cry.

The hillarious Seed of Corruption bug on PTR deserves a mention, I think.

They accidentally gave warlocks the hidden "detonation" spell part of Seed in their spellbooks; Seed of Corruption - Spells - World of Warcraft

Meaning, instant cast, 0 GCD, 0 mana cost, infinite range, AOE nuke.

Think about what this means for a second.

You can also cast it while you're dead.

/cast Seed of Corruption x 10

Farming/PVP goodiness! This will propably get fixed before tomorrow though :/

LOL

They will be in the next PTR patch.

It wasn't a bug, it was entirely deliberate. Think about it - what is the entire purpose of the current PTR push? To see whether the servers can handle the increased load caused by the network code changes. So, to stress test this properly, you want a server packed to the gills with people spamming instant cast spells.....

Pity the network changes weren't in yet

I hate bumping this thread, but there's no other place to put this:

Source: WoW Forums -> Holiday Maintenance Schedule 12/18/07

Running under some very reasonable assumptions, this means that 2.3.2 won't be pushed to live until January 8 (at the earliest), 2.4 won't be on the PTR until January 11 (at the earliest), and we may not see Sunwell until March.

Anyone know off the top of their head how long 2.3.2 has been on test so far? Also, is it okay to start complaining about the lack of raiding content yet? It's been 7ish months since BT and Hyjal were released...

Not really. As much as we all despise the "Casual vs. Hardcore" debate - fact is being on the more hardcore side you are faced with the wait and the patience that requires it. After the lift of attunements to SSC/TK - many are just starting to clear BT/Hyjal. So far so good. There's also a lot of gear to be had for BT/Hyjal guilds - many like to stay and perfect the fights, try alternate strategies, focus on loot - etc; the things they cant really do when the are pushing final bosses.

We've had this thread already. There's never been a wait this long, and I've argued that it's because Blizzard erred (though I didn't realize it until afterwards) by releasing so much of their raid content up-front instead of spacing it out more evenly (release t5 in April and t6 in August and people wouldn't be nearly as angsty today). But what's done is done. And it's not like complaining is going to make a difference in the development of an instance that surely is almost complete by now.

The mini-sprint wasn't just a case of throwing more abilities into the spell. Latency and client/server discrepancies in positioning made it so that even with a 20-80ms ping, it was often possible to shadowstep and immediately be out of melee range. It was an issue with the ability that definitely needed to be addressed.

It stands to reason that it's intended effect is basically what charge/intercept stun and feral charge root do, but without giving rogues yet another interrupt. I think they're moving in the right direction, but I personally prefer the control over damage play style, and prefer an arena setup where that role actually fits with my teammates.

If that was the goal, then why not have shadowstep ignore all movement impairing debuffs, until the sprint is over, similar to how intercept/charge work.

Because movement impairing effects have nothing to do with shadowstep mechanics. It's a teleport that happens instantly on server-side. The issue is the target isn't rooted or otherwise affected by shadowstep, meaning that under normal latency conditions they're not there anymore by the time you get there. The same thing happens with intercept often when the target resists the stun: you get there and they've left.

On re-reading your post I think you're actually talking about not being able to catch people afterwards because when snared the sprint isn't fast enough. Unless you have crippling poison on you you're still miving at faster-than-run speed, although outright snares like frost nova present a problem. Personally I think it should break movement-impairing effects anyways, since it's so similar to blink, but a 30-second escape artist is probably a tad too much for the developers. (yes, druids. they lose all energy when they try that; shadowstep costs 10)

That could be fixed by rooting (not stun) the target in place for 1 second instead of the movement speed buff.