Something that I always remind people is that the purpose of a resume is *not* to get a job -- the purpose of a resume is to get an *interview*. Its easy enough to confuse these things, but in fact they are quite different -- if you try to satisfy both masters your resume will end up with confused messaging at best, or worse, working at cross purposes; if you try to satisfy "get a job" then your resume is singing off-pitch.

So really, the ideal resume does not say "This person can do the job.", but says instead "This person is someone we should look at more closely (and nothing here says he *can't* do the job.)" A resume is just like the blurb on the back of the novel you might pick up while browsing the shelves of a bookstore -- the book might be great, but if the blurb fails to convince you that the cover is worth opening, its failed to do its job -- you put it back on the shelf and move on to the next book. In the same way, you don't try to convince a girl you fancy to marry you right away -- you're dead in the water if you did -- first you have to convince her you're someone worth a date.

Portfolios are the same way -- the portfolio together, as well as individual pieces within it, should be short, sweet, and too the point. They should demonstrate that you are capable, but not exhaustively so, and should never betray your weaknesses. They should always show your best work, and your best work only. It never works in your favor to misrepresent quantity as substitute for quality. Doing so dilutes your message -- in the event that you've managed to interest someone in exploring your portfolio, they probably only have time to look at one or two pieces; do you really want to take a chance that they land on something that's not your best work just to say you've got 10 demos instead of 3? I'd not even tempt odds to have 4 instead of 3, or 2 instead of 1.

For entry level programmers most of the time we don't bother looking at their site. First we look at (and typically discard) their application very quickly. Most get 10-20 seconds. A few get a full minute. This brings the pile of 50 or 100 down to a pile of 10 or less. Web sites are not a consideration at this phase.

What do you look for to decide? Entry level programmers will most likely only have a list of tech demos / small games, so having a bigger list might improve the chance of not being discarded so fast?