Will gravitational wave detectors be able to locate black hole collisions in time to view them? Can a single LIGO identify the general direction of a black hole collision? Could multiple observatories be able to accurately pinpoint the direction from them so that telescopes could quickly point in that direction to see the collision as it "is" happening?
 A: I'll break the question up into two parts: First, whether any type of real-time observation of a merger is possible.  (The answer is no, until we get space-based detectors.)  Second, whether LIGO is good enough to point detectors to where a merger has happened.  (The answer is no, but with Virgo's help, it can be done for very bright sources.)
First part: Real-time observation
Even the longest signals that LIGO will see are just tens of seconds long.  Simply finding the signal in the data usually takes longer than that, and the strongest part of the signal comes near the very end.  So even the "low-latency pipelines" don't tell us that there has been an event until ~30 seconds to several minutes after it ends.  But to get reasonably narrow window on the sky where we think the signal came from takes more like 15 minutes.  More detailed analysis required to narrow down the window can take many hours or even days.  So the prospects for real-time observations of any type of merger that LIGO detects are virtually nonexistent.  (There's good discussion here and in references therein.)
The fundamental reason for this is that LIGO's low-frequency sensitivity is quite poor, being dominated by seismic noise.  And since that's where binaries spend most of their time before merger, that's where we'd want to detect them to be able to see the merger.  Instead, they start accelerating into higher frequencies, so that by the time they're close to merger, they're really zipping through the frequencies that LIGO is sensitive to.
There is another class of detectors, however, that will be sensitive to low frequencies: space-based detectors.  The first one planned is LISA, and if you look at the plot shown here, you can see that it will be sensitive at frequencies 10,000 times lower than LIGO.  In fact, the "Resolvable galactic binaries" section on that plot is made up of a whole bunch of individual pairs that we should be able to identify uniquely, as they very slowly approach merger.  Since we'll have so much time to process the data from them, we should be able to pinpoint their locations very precisely, and even tell telescopes when and where they should point to see these events.
Second part: Pointing telescopes
Summary:


*

*Technically no

*In principle, for realistic systems maybe

*LIGO alone is almost certainly not good enough

*With a little help from Virgo (or future interferometers), it gets much better, but is still unlikely

*This actually happened for a pair of neutron stars three days after this question was asked (though it wasn't announced until two months later) and Virgo's help was crucial

*Nonetheless, what was seen from the neutron star merger was the afterglow, not the thing actually happening.



Well, first off, we don't necessarily expect to see black hole mergers.  Most of the mass in such a system is expected to be in the black holes (BHs), which are not directly visible.  In principle, there are two ways that visible/radio/x-ray/etc. telescopes could get information from such an event.  First, the mass would bend the light coming past the BHs — a process called gravitational lensing.  But the signal from this sort of thing is not easy to see, especially because this signal would be at its weakest when the gravitational-wave signal is the strongest.  Second, if there is matter very close to the BHs — like a disc or some gas — then the merger would change the dynamics of that matter, causing collisions that create electromagnetic emissions that telescopes could see.  The problem is that both of these are pretty weak.
Now, it so happens that LIGO alone is not very good at all at narrowing down the region of space in which an event occurs.  Maybe if a merger is really nearby, so there's much more signal than noise, we could do a really good job of pinpointing it.  But LIGO's two detectors were actually designed to be as closely aligned as possible given their locations.  This is good for proving that a signal is real, but bad for determining where it's coming from.  This is where Virgo (in Italy) comes in, and LIGO-India will come in.  Because they're in different places on the Earth, they're pointing in very different directions, so they're sensitive to waves coming from different places.  This helps tremendously in narrowing down search boxes for telescopes.
This picture shows the regions that LIGO was able to specify as likely origins for its sources, superimposed on a picture of the sky.  (The big band of stars that you see is the Milky Way, as we see it.)  Each detection so far is included, except for the most recent one.  You can see that most of them are really big regions, each having two separate pieces that are on opposite sides of the sky.  These are not much help for doing telescope searches because they account for a large fraction of the sky.  But two of the source, GW170814 and GW170817, are much smaller.  These are precisely the two detections for which Virgo was also turned on.  So you can see that Virgo is extremely helpful in narrowing down the region of sky over which telescopes have to search.

But still, we don't expect much light to be given off or strongly distorted by most BH mergers, so we don't expect to actually see them usually.  Maybe some gamma rays, but not much else.
However, the last detection on this image was not a BH merger.  GW170817 was the merger of two neutron stars.  They have plenty of matter, so there's plenty of opportunity to produce light, and there's no black hole horizon to escape (except possibly after the merger).  So we do expect light to be visible from these events.  And indeed, LIGO/Virgo was able to narrow down the search window enough that telescopes found all sorts of signals from this event.  Due to a couple problems that hopefully won't occur in other cases, astronomers weren't notified that this event had happened until 40 minutes after it ended, and weren't given a reasonably small patch of sky to search until 4.5 hours after it ended.  See here for more information.
A: Each LIGO facility is more similar to a microphone than a telescope. Microphones can only measure sound intensity, not the direction of the source. You can figure out the direction of a sound by using multiple microphones in multiple locations and recording the times that each microphone detects the sound. The order in which the microphones hear the signal and the time delay between the detections can reveal both the direction of the source and the speed of sound.
Since we only have two working LIGO facilities, we can only infer that the first detected black hole merger occurred somewhere in the southern sky since the Louisiana facility detected the signal before the Washington facility. The region can be narrowed by assuming gravity waves propagate at the speed of light and by comparing the difference in wave intensity due to the differing orientations of the two facilities.
With a third facility not in line with the other two, we could locate a signal to two possible points in the sky*. With four facilities that do not lie in a plane, we could pinpoint the direction of the source of the signal and directly measure the speed of the gravity wave. We're confident that the actual speed is the same as the speed of light, but measurements are always preferred.
* The two locations would be mirrored across the plane that contained the three facilities.
A: From the LIGO website on their "blind injections"
 (http://www.ligo.org/news/blind-injection.php), it does indeed look like it is possible to point telescopes in the general direction, but they haven't been able to do it accurately enough to get images yet.
Here's what they said about a blind injection event from 2010:

The detector network is capable of locating the source in the sky only crudely; it seemed to be coming from the constellation Canis Major (the "Big Dog") in the southern hemisphere (the event was dubbed "the Big Dog" shortly thereafter). They sent alerts to partners operating robotic optical telescopes in the southern hemisphere (ROTSE, TAROT, Skymapper, Zadko) and the Swift X-ray space telescope, all of which took images of the sky on that and/or subsequent days in the hope of capturing an optical or X-ray "afterglow".

(Thanks to Michael Seifert for his comment above on blind injections.)
