General Electric began marketing fluorescent tubes in 1938. Just a few years later, during WWII, Omega began designing lamphouses (cold light heads) using the new tubes. The most important reason for this was to meet the need for a more efficient light source for large format enlargers. Existing designs, such as the old Elwood and Saltzman enlargers, used parabolic reflectors and high wattage incandescent lamps. But they generated a great deal of heat, the lamps lasted only a few hours, and exposure times were still rather lengthy. Due to the size, weight, cost, and the need for multiple condenser sets for different formats, condenser heads for large format just weren't practical. The use of  fluorescent lamps allowed the design of light sources that produced a great deal of light without the heat, weight, and bulk. Omega was therefore able to meet the need of the military for 7x7 and 10x10 enlargers during the war.

Post war, there was another justification for the use of cold light heads. It was an era when black and white prints predominated over colour, and, in those days, there were no automatic print processors; all prints were still printed individually. Lab workers looked for any advantage that would increase productivity. The cold light diffusion lamphouse helped in that cause as it suppressed dust marks, fingerprints and scratches on negatives, reducing reprints and the amount of spotting and retouching necessary. Due to the Callier Effect, image contrast is higher with a condenser lamphouse than with a diffusion head.  Therefore, portrait photographers in particular appreciated the softer prints they got with a cold light head.

The introduction of 35mm Kodacolor film in 1958 resulted in much greater interest in colour prints, and with the development of colorheads and color print processors the cost of colour prints decreased and consumer preference quickly shifted away from B&W. As a result, the market for cold light heads diminished.  But it never disappeared entirely.  A relatively small group of users, particularly those involved in making fine black and white prints persisted, and a few small manufacturers continued to meet the needs of this specialty market.  Gradually, fed by the exaggerated claims of a few, a mystique seems to have grown around cold light heads.   

It has been suggested that prints made with cold light heads have an extended tonal range.  It may appear that that's true because of the lower contrast.  However, it must be remembered that a cold light head produces half to three-quarters of a grade less contrast than a condenser light source. If the proper grade of paper is used in both instances, the resulting prints will be virtually identical.  The reality is that the tonal range of the final print is limited far more by the quality of the negative and the tonal range that the photographic paper is capable of delivering than whether it is printed with a condenser or diffusion lamphouse.

Tungsten light includes an infrared component which is also absorbed by the silver halide crystals in the negative, causing the negative to gradually expand and eventually bulge or "pop."  Fluorescent light has no infrared component (hence the term "cold light") so it is true that users will not experience negative "popping" with a cold light head. However, the inference is that negative "popping" is a significant problem with condenser enlargers, which is not true.  It takes a considerable amount of time for the heat to build up in the negative to the extent that it pops, so it happens only infrequently. Installing heat absorbing glass in the condenser housing will also significantly reduce the possibility of negative "popping."

Does a cold light head eliminate "hot spots"?  Only if you have a poorly designed or faulty lamphouse now.  All enlargers should be checked from time to time for even light distribution.  One of the primary goals of any well designed lamphouse is to provide even illumination.  It's certainly not a property exclusive to cold light heads.

The assertion that one can get bigger enlargements because of a cold light head's lower profile is rather ridiculous.  That would only be true if the ceiling height were low enough that the top of your lamphouse touched it before the head was at the top of the column - and that's something that anyone setting up a darkroom should check beforehand.

The only claim that's completely true is that a cold light head will suppress dust and scratch marks.  But that's also true of any diffusion light source.  Even with a condenser lamphouse, proper processing, handling, and storage of negatives will prevent the problem in the first place.  Of course we may not always be printing our own negatives, so a diffusion light source may help cover the sloppy practices of others.

What you're not told is the problems inherent in using fluorescent tubes as an enlarger light source, and the lengths the manufacturers have had to go to address the problems. First of all, regular fluorescent tubes do not produce a full even spectrum of light. That's why colour photographs taken under fluorescent light typically exhibit an odd colour cast, usually green. And that's why normal flourescents don't work well with variable contrast papers. To get around this problem, the manufacturers of cold light heads eventually developed tubes with special (more expensive) phosphor coatings to achieve a more even spectrum that would work better with variable contrast paper.

The other inherent issue is that fluorescent tubes take some time to reach full operating temperature and full brightness. (Yes, they do require heat in order to work; they just don't radiate heat.) That caused inconsistent exposures unless the tubes were left on for the entire printing session. The first solution was to add a heater to the lamphouse to keep the tubes at operating temperature even when off. That helped, but did not fully resolve the problem. The next, more elaborate and costly, strategy was to employ "compensating" timers. These systems employed a photocell mounted in the lamphouse to actively monitor the actual light produced. The computer in the timer then automatically adjusted exposure times to keep exposures consistent despite variations in light output.

While a cold light head can be a perfectly fine light source for making black and white prints, it is not a "magic box", nor is it essential for making quality prints, as some would have you believe. It is simply one type of diffusion light source.