I have actually always been amazed at how long this is taking. I realize that the Source Engine is probably harder to map for than the Unreal Engine 2 was, but still.
One team (they changed the roster a bit, but it was still the same guys) released two and a half major mission pack/campaigns (plus a bunch of maps on the side) in ~5 years. These were original levels (with damned fine level design), new guns, and even a lot of new tweaks to the engine/gameplay. These campaigns, for those interested: http://rednemesis.beyondunreal.com/
How long has Black Mesa been in development? 5 years? And they are doing remakes of levels, so the main problem is just making the assets and geometry. HL1 wasn't THAT long, was it? And it isn't like the environment changed too many times (Outside, Xen, Sewer, Offices, Lab), and I think the game used a lot of reuse anyway.
Don't get me wrong, I am looking forward to it. I just wonder if they will be able to justify how long it took.

Hey, doesn't that mean DukeNico wins that bet with that other guy over the release dates?

Looking at the screenshots, I'm not impressed. Those levels are a mess. They look like they could have been taken straight out of Quake 3.

They were for Unreal Tournament, which basically came out at the same time as Quake 3 :p
And actually, some of the later levels of 7 Bullets were breathtaking and probably spanked a lot of the 2k4 maps.
But I digress. My main point is that I just don't know how it is taking so long for Black Mesa. Are Source and Hammer really that hard to use?

probably a lot of drama.

Ugh. That shows how good my reading comprehension is. I would have sworn you said Unreal Tournament 3 (which wouldn't have lined up time wise anyways, as UT3 definitely hasn't been out 5 years).
My apologies.

That and rebuilding an entire game with all its assets from scratch is not as easy as you make it sound for a bunch of part-time amateurs.

If you're talking about The Nameless Mod, that mod took 7 years to finish.
In my opinion it was worth the wait though, that mod can kick most full games in the square in the balls and make fun of their moms. You can almost feel the time spent on it the moment you start playing.
Though it also has to be said that it could have been done less than half the time if they had a paid staff working on them regularly.
Again, these things can go really slowly simply because the people working on them are not nearly as well coordinated as a paid team of developers. Even the little problems that wouldn't cost a professional developer a day can set you back by several months.

Red Nemesis were part-time amateurs (okay, one or two of them got real jobs, but I am pretty sure that was after Xidia Gold :p).
Don't get me wrong, it is a lot of work. But you have to admit: Making things pretty is a pretty simple thing to do ("We need new textures. You, you, and you. Go remake these textures in hi-res with prettiness") when compared to actually balancing gameplay and level design (something Valve already did).
Either they are completely remaking the game (which is not what they claim to be doing. Everything I have read, on their site, suggests that they are effectively just duplicating the levels, but prettier), which will probably piss off all the people who have been following this, or Hammer and Source are just really hard to use (I grew up on subtractive level design, so I can't really comment too much).

Its not just a matter of "making things pretty" though (which is actually far more difficult than you make it sound, especially to do it well). They had to re-create every model and texture from scratch (except for those that they could have re-used from HL2, like the crowbar and Freeman's hands), then re-create each level from the original game, matching it precisely, then get things like the audio and lipsynching working, then I'm sure there were other things that had to be done that I'm not even thinking of right now. This was not a simple process by any stretch of the imagination. Could it have been done in a shorter amount of time? Probably... if they were working on it full time, every day, at least 5 days a week, but they weren't. This was a part-time hobby for a relative handful of people working separately in different parts of the world. The fact that are near finishing at all is a minor miracle.

Oh, I fully agree that the fact that they are fishing at all is a miracle. But that is more because of my low-expectations and general experience with modding teams (especially ones that are, essentially, cover bands). :p
Again though, any team would have to do those activities. And my point was more that you don't need nine people to work on the same texture, that work can easily be assigned on a person-by-person basis.
And considering that they have a team with a size that rivals some smaller dev teams, they definitely have enough people to assign work to. Sure the time spent by each person is somewhat low (adding to the overall development time), but they can pull off some throughput very easily.
And you yourself point out: they are (probably) matching the levels precisely.
I am going to pretend that you have painted something before (hell, let's pretend that I have :p). It is much harder to paint your own picture than it is to duplicate another person's picture. Even though you have to make your own strokes and the like, things like color and perspective have already been thought out.
So, at he very least, I think that should even out a lot of the asset recreation (which, considering the reuse in HL1 isn't as massive an undertaking as you are describing. Especially if Portal's assets are available, because they provided a lot of office furniture and the like. And HL2 already came with some assets that can be used for equipment for the test chamber and the sewers).

Not any team takes their work as seriously as these guys do, which clearly sets it apart of other mods. I don't get your point, each member works in his own specialization which can be level designer, prop modeller, programmer, etc... If there comes the desire to mutually add to each others work this will only benefit the quality of the project and isn't even lost development time.
I think you are underestimating the task at hand (that's even an understatement). Managing such a big team gets really messy very quickly. You can't count on each one of them to be available at any given time, you have to expect the unexpected at all times and if an important coremember decides to quit it will slow the progress considerably down.
You're repeating "easy" and I don't see how this is easy at all! As Project Leader you need to find ways to motivate your fellow workers to keep both the Developers from leaving and the mod from dying. They forced a deadline on them to have something to look forward to to get it done, on the other hand they are not willing to sacrifice their work just to meet the imposed date.
Contrary to what you could think, they are not matching the levels precisely to the originals! They're building it all up from SCRATCH. Improving, fixing, giving a meaning to the levels by documenting themselves with architectures, modern facilities, hundred if not thousands of reference shots they need to dig up or take themselves in order to do a believable world.
Some whole sections had to be redone entirely to match the current level of quality, gameplay mechanics, AI, scenes... This sets you back several months already!
It's actually the small details like something as trivial as assets and props that cost a considerable amount of time to get done right.
They are using almost nothing from the shared content, of the top of my head the only stuff they're keeping is the headcrab, vortigaunt and crow. This has the advantage for you that you don't need to own one precise game but any Source engine powered game that comes with the SDK.

My point with regard to the large team was more along the lines of being able to spread work out. It is the old latency Vs throughput argument. If you assign work properly, you don't need person A to wait for person B. Person A can make his plant and person B can make her plant, and at the end of the stage persons A and B can tweak the other plant to make them fit together more.
And does anyone have any official word on what they are doing? Their site suggests that they are just remaking the levels (same floorplan, but different aesthetics), and most of the people I have talked to (who follow this) seem to think the same thing. But you just suggested that they are doing a massive remake with different floorplans and likely even different gameplay. If so, I can see how they are actually finishing. Rather than being a cover band, they are just a band pretending to be a cover band :p
All I am saying is that the thing which should take the longest (level design and balance) is already handled for them. That just leaves pretty stuff, which I just don't see how it can take this long, especially with this many people involved. There are only so many assets in the game (again, reuse).

You are simply underestimating (way underestimating) the ease of what they did have to do. Sure the levels may have been mapped out in the original game, but almost none of it had been created at all in Source. They had to re-make almost everything. You seem to think that the level design part is the hardest part of this whole process, I submit that you are completely mistaken on this point. Conceptual work can take a long time to accomplish, but actually taking something from a concept to a finished product is well more than half, if not 80-90% of the development process. Add to that the complexity of a team doing it all over the internet in their spare time, like I said before it is a miracle they have gotten this far int the time they have spent on it.