Slashdot Mirror


User: demultiplexer

demultiplexer's activity in the archive.

Stories
0
Comments
5
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 5

  1. Re:Post megapack on AOL Tries New Tactic to Keep Customers · · Score: 1

    I really don't buy this "He's doing this so he can pay rent [...] ask to talk to a supervisor" at all. You work for this company, in fact you are the face of this company. Therefore you both act and receive on behalf of the company. The problem is entirely your own - you're angry because you have been taking customer complaints and attacks personally. Well, as you said, you're just "a peon". The customer is angry with the way your script treats him, with your company, not with you personally. When he starts ranting, he is ranting against your company, and he has every, absolutely every right to do so, because you are representing it. I cannot stand staff who exhibit this "it's not my problem, I just work here" mentality. It is the bane of decent customer service. It is your problem because you work here.

  2. Re:Is Java finished? on Java vs .NET · · Score: 1

    There you are. You cannot modify and distribute, so open source it ain't. But hey, it's better than nothing.

  3. Re:Since the author didnt mention it... on Design Patterns · · Score: 2

    It may be the first and oldest (software) pattern book around, but it still is the very best I've read. Consider this. The book introduces 23 patterns. In the ten years elapsed since its appearance, perhaps half a dozen more patterns have been formulated that can hold their own beside them. Others are derivative, have a very specific application domain, or are about something entirely different altogether (from XML patterns through organisational patterns to anti-patterns).

    If you'd write the book today, you would probably use different programming languages but really, the concepts may be complex but the examples aren't. If you are proficient in one or two modern OO languages I doubt you'll have a hard time deciphering them.

    Treat yourself. Buy the book. Keep it in your bag and next to your bed. It will stretch your mind, clear up your thinking, and give you a new vocabulary.

    - Peter

  4. Re:Trademark Rules on JavaRanch gets Cease And Desist From Sun · · Score: 1

    It is true that if Sun doesn't enforce their trademark, it will become un-enforceable in the end. So we are asking them to sell us a license (preferably for a symbolic fee). That way, JavaRanch can remain JavaRanch without weakening Sun's claims on the Java name. No joy so far.

    The infuriating thing is not that they try to take care of their Java(tm) trademark, but that they are being so inflexible and non-cooperative about it.

  5. Re:Even worse! Incosistent math! on Mathematical Analysis of Gnutella · · Score: 1

    Indeed, the values for N and T are silently increased and the figures you get for the defaults are much more reasonable.

    Thanks to that same N and T, Gnutella has an inherent limited visibility of 484 nodes by default. Use the author's own queries/user figures, we find that any area of the network is hit by 0.3 queries/second, a long way short of the quoted figure; the traffic flowing through it is proportionally lower at 74KB/s. As a consequence there cannot be "hot spots" in the network, the traffic is smoothed out.

    The total calculated traffic of 2.4MB/s (N=4, T=5) is absolute peanuts when it is spread over the entire Internet. The fact that a Gnutella net is rarely balanced tends to decrease, not increase, the bandwidth consumed. So in real life you won't reach even that.

    What's more, with an increasing number of users, the amount of traffic generated increases linearly. This isn't bad scalability - that is about as good as it gets! Where Gnutella scales badly is when network visibility (N and T) increases. That has nothing to do with the number of users though.

    - Peter