Slashdot Mirror

If it isn't a problem for your wallet, the future is now:

http://www.codinghorror.com/blog/archives/000927.h tml

I think what he means regarding bandwidth being a 'running cost,' is that a low-level consumer ISP has to buy their connectivity from somewhere.

Let me illustrate how it works. I'll use the example of an old dialup ISP, because it's simpler. It used to be, back in the mid-90s, that anybody who wanted to could become a local ISP. You'd go and get a bunch of phone lines, a bank of modems, and a T1 or fractional-T1. Customers would call in, connect to your modems, and then you'd push their packets out the T1 to your ISP, generally a regional carrier, who would take them to a higher-level ISP ... who would be big enough to engage in traffic peering with other major networks.

Basically, the money flows up in this scenario. Customers pay you (the rinky-dink local ISP), and you pay your regional ISP for the T1 line, and the regional ISP pays to connect to the higher-tier network. (The highest-tier networks, the Tier 1's, don't pay anybody -- they basically have gentlemen's agreements and pass traffic to each other as peers and equals; hence "peering." But there's not much peering below the Tier 1/2 level, since the traffic is too unidirectional.)

Now, the customers pay a basically flat rate per month ($20/mo or whatever) regardless of how much they use, within reason. But the ISPs usually aren't so lucky. When you buy a T1/T3/OC-x or other significant connection, you pay for the connection itself, and for a certain amount of bandwidth. In many cases (particularly on the faster ones), the amount of transfer that's included is less than 24/7 saturation -- in fact, you'd want it to be less than saturation, because 90% of the time you won't be saturating it. You want to have some burst capacity for the 6PM hour when everyone comes home from work and checks their email, but you don't want to pay for that much bandwidth all the time.

So the cost of your upstream bandwidth is variable, not fixed.* If your customers start using a lot of traffic, suddenly you as the ISP end up paying a lot more to your provider for bandwidth. That cuts directly into your profit margin.

Now ... moving up from dialup ISPs to a high-speed provider like Comcast, they have problems that are even more severe, because they oversell their networks even more severely than dialup ISPs do. While a dialup ISP can only ever have (modem speed)*(number of modems) as a maximum, a broadband ISP could conceivably have (connection speed)*(number of customers). If everyone on the network decided to pull down 6Mb/s, everything would grind to a halt -- they don't have the capacity to the upstream providers for that kind of bandwidth. And even if they did have the physical capacity, they couldn't afford it with each customer only paying $60 a month. (Transfer isn't exactly cheap.)

However, depending on how a broadband ISP like Comcast has their network set up, they could probably rig it so that packets that go from one Comcast customer to another are "free," in that they don't go through a higher-tier provider. This requires that Comcast have its own internal network connecting all the neighborhood nodes, instead of just wiring them all directly to upstream ISPs without interconnection. I'm not sure whether they do this or not. I have a suspicion that they might, because I've heard that they're not throttling Bittorrent connections from one Comcast user to another. This makes sense, if those packets cost Comcast less than one going from a Comcast user to a Speakeasy user.

ISPs hate P2P applications like Bittorrent because they're bandwidth hogs, but more specifically because they result in a lot of extra traffic being pushed through their Internet gateways, which means more bandwidth costs for them. If you could set up a P2P app that only peered with other nodes on the same subnet or IP address range, the ISP's might care about them a whole lot less (though it depends how they have th

That's just as misguided.

English: Engineers need it more than most. Expressing yourself effectively is vital or else you end up losing money or possibly croaking people because your audience paid attention to the wrong part of your PPT.

Diversity?: This class doesn't exist. You made it up. Besides, engineering classes are the most diverse on any campus. You meet geeks and nerds from 5 different continents!

Art history: Do you want to build this?
http://www.codinghorror.com/blog/images/butterfly_ bridge.jpg
or this?
http://clockhistory.com/setclocks/orig_factory_pho tos/images/factory_1912.jpg

How about this?
http://www.subrew.com/library/yugo/yugo_front.jpg
or this?
http://www.diseno-art.com/images/lotus_elise_111r. jpg

The author was NOT calling for a network where all information is freely available to everyone, simply an open framework within which people can network as they please.

Really? There's a lot of specific examples in the article that pretty much say exactly that, AFAICS. The stuff about interoperability and standards is all very nice, but how exactly do you think those are going to work without making the data itself open to some extent as well? What you're basically talking about, if you don't have that, is just the usual "wouldn't single sign-on be nice?" musing.

Methinks a few too many people in this discussion have read this one article and missed the wider context. This is hardly the first post using the walled garden metaphor. Bloggers were talking about this several weeks ago; see this blog post at Coding Horror and the various things linked around it, for example.

You might have also given a link to the slightly less inflammatory titled folluw-up post to "What's Wrong With Apple's Font Rendering", called:

Font Rendering: Respecting The Pixel Grid
I've finally determined What's Wrong With Apple's Font Rendering. As it turns out, there actually wasn't anything wrong with Apple's font rendering, per se. Apple simply chose a different font rendering philosophy, as Joel Spolsky explains:... (link to article)

Rather good and concise explanation of the different strategies of font-rendering.

"Mac users will continue to see the Internet as it was intended"

What's Wrong With Apple's Font Rendering?
Welcome to the blurry, but fast, browser...
Apple and Microsoft have always disagreed in how to display fonts on computer displays...

Sorry, thats a lie: See here an average of around 5.7

http://www.codinghorror.com/blog/archives/000818.h tml

Another example here, with more text. Also a very helpful close-up view (in case you can't be bothered opening up an image editor) of text rendered in Safari and in IE 7 (no Firefox screenshots).

Joel's spot-on about the difference in ethos. I'm not sure which I prefer, aesthetics or effortless legibility. Maybe one day I can have a 150-200 dpi monitor and then I won't need to care.

Another example here, with more text. Also a very helpful close-up view (in case you can't be bothered opening up an image editor) of text rendered in Safari and in IE 7 (no Firefox screenshots).

Joel's spot-on about the difference in ethos. I'm not sure which I prefer, aesthetics or effortless legibility. Maybe one day I can have a 150-200 dpi monitor and then I won't need to care.

Another example here, with more text. Also a very helpful close-up view (in case you can't be bothered opening up an image editor) of text rendered in Safari and in IE 7 (no Firefox screenshots).

Joel's spot-on about the difference in ethos. I'm not sure which I prefer, aesthetics or effortless legibility. Maybe one day I can have a 150-200 dpi monitor and then I won't need to care.

AFAIK this technique was pioneered by Jeff Atwood with his "ORANGE" captcha.

--Suse 7.3 DVD install was actually really good back in the day. Reiserfs and LVM supported out of the box. (22 October 2001)

--XP was originally released in Oct 2001, as well.

See:
http://www.codinghorror.com/blog/archives/000646.h tml

Mmmm... not really, at least not in XP. You can turn OFF automatic updates, but you can't disable that dialog from there. The same dialog *I think* appears if you do manual updates, though maybe not through IE. (If you have auto updates set to just inform you or download and inform it definitely still brings it up.)

This is the link you really want to turn off the dialog if you still want auto updates on.

This worries me. I don't think the street's interests match up with open source communities or the private companies that make a living off of them. The street likes bold moves for short term gains. OSS demands slow(usually) organic growth and consensus building. A MySQL controlled by arms-length investors might throw an error, how will Mickos catch?

This http://www.codinghorror.com/blog/archives/000842.h tml comes to mind.

If they offer a second class of shares(dividends but no voice) then maybe it will work, but I'm not sure there's a point in it.

Why is Ruby suddenly the big buzz these days?

My internets inform me it's a high level server-side language (python for websites).

I found an article on coding horror in the depths of my RSS reader talking about how Twitter are having no end of problems scaling their system, as they're using Ruby and it just isn't made for big, high traffic sites.

If the main reason for writing server side software is web based applications, or at least dynamic content, isn't a huge factor in this how well it scales?

Nobody makes sites to be used by 20 people.

Finally, if anybody can explain it's popularity to me, should I learn it?

I'm currently doing freelance web dev mostly in PHP, would it be useful?

In my spare time I'm writing a AJAX web app with PHP back-end at the moment and it's mostly for my personal use (task tracking from anywhere), is Ruby good here with the limited audience the site'll have?

Why is Ruby suddenly the big buzz these days? My internets inform me it's a high level server-side language (python for websites). Isn't the most important thing in a server side language fast performance rather than quick development time? I found an article on coding horror in the depths of my RSS reader talking about how Twitter are having no end of problems scaling their system, as they're using Ruby and it just isn't made for big, high traffic sites.
If the main reason for writing server side software is web based applications, or at least dynamic content, isn't a huge factor in this how well it scales? Nobody makes sites to be used by 20 people.

Finally, if anybody can explain it's popularity to me, should I learn it? I'm currently doing freelance web dev mostly in PHP, would it be useful? How? In my spare time I'm writing a AJAX web app with PHP back-end at the moment and it's mostly for my personal use (task tracking from anywhere), is Ruby good here with the limited audience the site'll have?

Your computer doesn't necessarily use 400 watts all the time. It might use mebbe 40-50w in idle, AKA, no 3D apps running, nothing using too much power.

Here are a few sites with some more info:
http://windows.uwaterloo.ca/Hardware/PC_Power_Cons umption.asp
http://www.codinghorror.com/blog/archives/000662.h tml

This guy has some good points and it's worth a read. Title of the article is "Pick a License, Any License"

Would some study results from a research company help persuade your bosses?

Microsoft researchers haven't perfected the genie, but they've found a tool that can increase your productivity by 9 to 50 percent and make your work day easier. And you can begin using it right away. The researchers conducted user studies that proved the effectiveness of adding a second or even third monitor to your workstation, creating a wide-screen effect. In addition, they found out how the operating system needs to change to accommodate a larger screen area.
http://research.microsoft.com/displayarticle.aspx? id=433

Or how about this one?

A systematic study conducted by NEC-MitsubishiOpen link in new window, ATI TechnologiesOpen link in new window and the University of UtahOpen link in new window has concluded that the use of multiple monitors in the workplace increases productivity.

Granted, you have to take the results of research with a grain of salt when the team is headed by a company that makes monitors, but still...

And if they're the kind of folks that like anecdotal evidence, just send them here and here.

hi,

yes, two monitors can make you more productive. there has already been an article on slashdot
about this:

http://developers.slashdot.org/article.pl?sid=03/1 0/09/137232

and there are also:

http://www.netlobo.com/dual_monitor_productivity.h tml
http://www.codinghorror.com/blog/archives/000012.h tml
http://lifehacker.com/software/dual-monitor/dual-m onitors-increase-productivity-168488.php

Perhaps your experience with much of the system memory being used is due to SuperFetch. See the article below for more information. Bottom line, however, is a.) Vista may be using your system more efficiently and b.) if you don't like SuperFetch, it's easy enough to turn off.
http://www.codinghorror.com/blog/archives/000688.h tml

Battery life would be radically better

http://www.codinghorror.com/blog/archives/000688.h tml

Vista uses all of your RAM all of the time. It lets go what it needs to for applications but uses the RAM that normally goes "unused" for caching. Every OS should do this.

And apparently they manage to weed out 80% of applicants through just that test - people who might be good at a memory test of "do algorithm X", "virtual bass class Y" and so on, but would be crap at work having to think on their feet.

If you are a programmer, you have to have specs. If you are an analyst or architect (or a consultant), you must engage in requirements gathering and business analysis, finding out what your client wants and how they do their work right now. If you work as an employee, consult your job description. Note that requirements gathering is a specialised process, and is very difficult if you have to design software for professional or scientific use (such as accountancy software or engineering applications), so it would be a good idea to cooperate with a specialist (e.g. an accountant) or read some introductory books about accountancy so that you can understand what your clients does right now. Oh, and good luck!

The default config used over 400MB on bootup and I find this appalling ...

It's called SuperFetch and it puts your RAM to good use when it's not needed for anything else. You can disable it if you'd like.

slmgr -rearm

(Addendum)

Here are two more links with yet more evidence that O/R mappers are very far from problem-free:

http://blogs.tedneward.com/2006/06/26/The+Vietnam+ Of+Computer+Science.aspx

http://www.codinghorror.com/mtype/mt-comments-rena med.cgi?entry_id=621

Basically, these tools are becomming more complicated than the very database they are trying to hide/wrap! Keeping app developer's inside an OO view of things is becomming more important than developer productivity and simplicity. It smells of W-style "stay the cOOurse" zealotry.

The Sugar UI looks very interesting. Jeff Atwood wrote about this just the other day. Its kind of refreshing to see a project like this break new ground in other areas besides hardware. Unfortunately there have been rumors that there has been zero usability testing so far. And yes, you can test it out for yourself, but maybe it would be even better to let your kid try it and email them the results.

> The organization makes choices for them

You mean like this?

http://www.codinghorror.com/blog/images/software_e ngineering_explained.gif

Is there a 'So true it hurts' mod?

I don't "do" C/R. If I mail you and you challenge me I hit delete, because, as Dan Quinlan put it: "C/R is the ultimate email diss. By using it you are saying, 'my time is more important than yours.'"

"take up the whole fucking screen with this window" stems partly from being stuck in the Windows singletasking frame of mind

It's not a "frame of mind." Single-tasking is what humans are optimized for. There's a reason IQ goes down when you try to do more than one thing at a time.

I have a Mac as my primary computer (using itto write this) and I generally love it, but sometimes I'd rather fill the screen so that there's nothing else on it but the task at hand, just like I regularly remove all distractions from my physical desktop.

Fortunately I found Stoplight. The very fact of its existence AND the fact that there's not an equivalent product for Windows ought to tell you something.

That can happen when you have more than 1GB of RAM. That happened when I upgraded to 2GB of RAM in my system. I did a quick Google search, and found that there is actually a Microsoft released hotfix for the problem.

Long story short, VSS is one of the worst piece of source-control software ever. This is a quite beaten to death horse, so here are 3 articles on the subject: Visual SourceSafe Version Control: Unsafe at any Speed?, Visual SourceSafe: Microsoft's Source Destruction System and Source Control: Anything But SourceSafe.

The images don't seem to be loading on the site, but why do I get the feeling it's going to be somewhat reminiscent of this?

With a 2 ghz processor and 1 gb RAM, at Idle I was pushing 70% physical RAM usage and a constant 10% load on the processor.

One thing that jumps out at me is your criticism of Microsoft for not following their own Guidelines. However, I don't think you've ever seen new versions of MSFT software running on Vista. There's no reason to think that this behavior, such as shortcut naming, etc, hasn't been brought inline with these reccomendations.

Shortcuts, from what I've seen, have been rebranded.

For instance, the Windows name is now all over Windows applets, even the ones that used to be known with unqualified names, such as Calculator.

• http://www.codinghorror.com/blog/archives/000555.h tml
• http://geekswithblogs.net/cbreisch/archive/2006/04 /11/74818.aspx
• http://www.neowin.net/index.php?act=view&id=34631

Well, yes, you need people who are able to work in a group. An interview, however, is much different than such groups. In the group, you've already been accepted, likely have some familiarity with the other members, and have further understanding of the processes, values, and expectations of those involved. In an interview, you typically do not know the interviewer, are in the spotlight, and have not yet been accepted. Much like the interaction between two people on a first date is stilted and awkward, the interaction between married couples is much more fluid and indicative of the members' everyday nature. I also didn't mean to say that all coders are socially inept, just more than in the population at large.
Being a business, as you say, it's ultimately your choice as to what qualities you evaluate in potential employees. I can't argue against that any more than I can argue that Arrested Development is a hilarious show. I don't really have significant experience in being responsible for finding staff, and doing so efficiently. I have no business experience, per se, and I appreciate (literally and figuratively) those who do. I do know programming, however, and I have experienced and been witness to many speciously good interviewees with strong social skills gain employment where more deeply talented but awkward people have failed. I'm sure the businesses absorb whoever they get, and things still proceed profitably. Big companies can do this.
Along with gender, weight, age, appearance, and race bias in interviews, social skills are just one of many areas which obscure pure merit. I'm surprised more companies don't employ rigorous application tests and psychological evaluations to replace the caprice of the varied interviewers. It would potentially save time and money, while providing a more standard yardstick to measure applicants.
Again, I'm not addressing this to you specifically. I read your web page, and you are a card-carrying computer geek. I'm just surprised by the relative lack of thought and resources applied to such an important aspect of building a business.

I've read two really good items on the subject of estimating software schedules. The first is Painless Software Schedules by Joel Spolsky, the Joel in Joel on Software. It's a quick read, and a lot of the comments here are giving the same advice all spread out. Even more useful is Waltzing With Bears by Tom DeMarco (ISBN 0932633609) (very talented author, I strongly recommend Peopleware to everyone), which is about managing risk on software projects, especially as it relates to time. This is one of the fundamental errors in the question "How long will it take?"--the inquirer wants an exact amount of time, so if you say it will take four hours, it should take exactly four hours. The problem is, you may luck out--someone may have had that feature in the program once already, and it was removed, so all you need to do is call some fully-tested code from a different place, or there may be high coupling, so that what looks like a really simple, straightforward change is insanely hard. Those two factors put together means giving an estimate should be something like "No more than forty hours, 75% probability of finishing within twenty-four hours, 50% chance of finishing within six hours, 25% of finishing within four hours, no quicker than an hour." As part of Waltzing With Bears, Tom DeMarco (and I assume others) put together a Riskology spreadsheet, intended to allow you to estimate schedule probability curves, which allows combining multiple probability curves to get an estimate more like "No faster than eight months, 75% within six and a half months, 50% within five months, 25% within ten weeks, no faster than four weeks." And always make sure the first number people hear is the worst case scenario--that's the one they're going to remember.

Other reading:
Coding Horror, How Long Would It Take If EVERYTHING Went Wrong?
Software Estimation: Demystifying the Black Art by Steve McConnell
Google on Estimating Software Projects

Speaking just to the power-savings benefit of using a mobile CPU in a home system, unless you are running a home server, the best way to conserve power in any PC would be to turn it off.

This link notes one person's cost of leaving systems powered up, a little more than $23.00 USD per month.

So boot it up each morning, make the coffee and toast, and by the time you sit back down in front of Unbuntu, Mandriva, BSD, SUSE or that friendly chair-throwing group from Redmond, you can feel better about not burning up more oil to play solitaire (YOU'RE FIRED).

Most people can get by with a 250 watt power supply these days. A hard drive takes around 25 watts to spin up but after that it cruises along pulling only 4 to 6. Check this out.. http://www.codinghorror.com/blog/archives/000353.h tml

Find some other way to compensate for your lack of whatever instead of burning more energy just because it's "bigger-than-yours".

Yes! I found it! What dogs hear

I was so irritated after posting my last message that I did some digging on the problem. I found these registry changes, although I haven't tried them yet..

[HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\Wi ndows\WindowsUpdate\AU]
"RebootRelaunchTimeoutEnabled"=dword:00000000
"NoAutoRebootWithLoggedOnUsers"=dword:00000001

Courtesy of: http://www.codinghorror.com/blog/archives/000294.h tml