Slashdot Mirror

I'll tell you why, no wait. I wont. Maybe Paul can tell you why. The following is from his General FAQ
"Why is the text on your site so narrow? It wastes screen space.
The aim of web design is not to use all available screen space. It is legibility. Text is most legible with at most 60-70 characters per line. On computer screens, you don't want to go much over 60."

Then I had to actually interact with Yahoo! Stores. Is it just me or is the usability lousy? Is it just me or does the sales process suck? Maybe whomever ported it broke it?

He can write. Maybe he can code. His approach to user interaction blows.

How do you know? Lisp has been around for a while, and it's not dead, yet. Some Lispers are working on a language called Arc, which they hope will last a hundred years. On another front, perhaps Parrot or .NET will provide a stable base that will allow languages to evolve, while remaining compatible.

That said, I don't think it's necessary for a long-lived software project to use one language, exclusively. Standard interfaces can commoditize the language, to some extent.

It is not easy to design a 100year old language

Still, on a more serious note, one has to wonder what that means over the long term. One needs the ability to read and read effectively to digest technical material like Comp Sci, or to communicate with others. If fewer Americans read fiction, the foundation upon which most comprehension is built, I can perceive negative consequences over the long term. A less informed, less scientifically-inclined population could lead to eventual structural deficiencies in society.

Then again, maybe the problem also lies in the school system itself. As Paul Graham points out, most English teachers have an extraordinary ability to deaden material. The novel Election by Tom Perrotta goes into this high school malaise, albeit in a subtle way.

Indeed; few other first world countries have such a stigma against using your mind for more noble functions. Having extended experience with computers and programming brings a label of 'pathetic nerds', while useless sports knowledge and statistics is often considered par for the course.

There's an interesting
article written by Paul Graham about this problem.

Is UML useful? How about books on secure programming? Are design patterns a good tool? Will learning things like assembly or Lisp teach me anything useful?

All of these (well, I don't know UML, and I am only planning on learning patterns) are useful. I'd especially reccomend learning LISP and assembly. LISP will teach you new methods of abstraction above and beyond OOP, as well as "programmable programming language" ideas that can be used to understand language-writing with less pain. (For more on why LISP is good, see Paul Graham's site.

Assembly language is very useful because you find out how the computer works at a low level. When you know that, it will help you make your higher level language programming more efficient since you'll have a better idea of how it will work once it is compiled or interpreted. That's why Knuth's "The Art of Programming" has its examples written in an imaginary assembly language.

It's more likely that programmers will just use languages that waste machine time but save programmer time.

On the other hand the death of C or C++ has been predicted for about 20 years now, and it's still pretty popular. So I don't think large C packages will go out of style any time soon.

LISP tried to replace flow control with function calls, yet this didn't become the dominant programming model. People who swear by it end up using functions like apply that implement flow control statements like foreach.

The misanthropes that took over comp.lang.lisp are pathetic. I've never seen a techical discussion group that hostile and defensive.

Common Lisp fossilized sometime back in the Reagan Administration and has since lost almost all ability to improve.

As a result, the vast majority of former users have abandoned it and those who remain almost have to take a position that there is no further NEED for improvement except in trivial ways (more libraries, more "complete" implementations, etc.) that, if you think about it, are merely restatements of the "nothing needs to be improved" notion.

Arc is announcementware. It has shown no signs of life since its first few weeks.

There are several dozen different Schemes, all incompatible, with an average of maybe 1.1 implementers each.

Of course, if you really want a one-implementation, brand-spanking-new (no compatibilities here, sir!) Lisp, then by all means check out newLISP.

What we really need is a meta programming language of which all the other programming languages are special cases.

For a book that discusses this concept at some length, I recommend On Lisp by Paul Graham.

Beating the averages

Both are amazingly good.

Beating the averages

Both are amazingly good.

• the Hundred-Year Language
• Succintness is Power
• Five Questions about Language Design
• What Languages Fix

• the Hundred-Year Language
• Succintness is Power
• Five Questions about Language Design
• What Languages Fix

• the Hundred-Year Language
• Succintness is Power
• Five Questions about Language Design
• What Languages Fix

• the Hundred-Year Language
• Succintness is Power
• Five Questions about Language Design
• What Languages Fix

• the Hundred-Year Language
• Succintness is Power
• Five Questions about Language Design
• What Languages Fix

The ability to concentrate on one thing (attention span) is highly useful. I learned to apply it to schoolwork, so I was quite successful in high school. Find something that interests you, that you can obsess about. You can use that to train your attention span to more conscious control.

In the short run, I'd suggest that you read Paul Graham's essays Why Nerds Are Unpopular and Hackers And Painters, from his book that was reviewed on Slashdot this past Tuesday. The first has some insight on why you probably were not interested in school, and the second speaks somewhat to your gifts.

The ability to concentrate on one thing (attention span) is highly useful. I learned to apply it to schoolwork, so I was quite successful in high school. Find something that interests you, that you can obsess about. You can use that to train your attention span to more conscious control.

In the short run, I'd suggest that you read Paul Graham's essays Why Nerds Are Unpopular and Hackers And Painters, from his book that was reviewed on Slashdot this past Tuesday. The first has some insight on why you probably were not interested in school, and the second speaks somewhat to your gifts.

At last, an explanation for the pages of drivel pg recently published as "What You Can't Say". For such a smart guy, I was agape with confusion as to how pg had gotten himself into the absurd position of arguing that heresy is "cool", and if you are not a heretic then "...Odds are you just think whatever you're told" - it was like reading the rants of some eloquent teenager, full of childish angst and rage toward authority.

The word heresy means to choose or to pick out, and libertarianism, which chooses the rights of the individual over the rights of the collective whole, is an apt example of how pg's much-advised heretical thought ends in useless, unoriginal crackpottery. Slowly, a heretic's monomania warps his mind, preventing him from seeing things in a universal sense, according to the whole, and I think it highly unfortunate to see pg slowly manifest these symptoms.

In the end, as G. K. Chesterton said, there is nothing more boring than a heretic, and pg has become exceedingly boring of late (of course, not quite to the standard of Mr. E.S. Raymond or Mr. R.M. Stallman), as evidenced by the content of his essays, and the time he is wasting by implementing another pointless dialect of ANSI Common Lisp. It would certainly benefit us all if there were at least one person with voice in the open source world who thought in a more catholic manner; then perhaps we would see some progress instead of good people being wasted, thinking unoriginal thoughts, and re-implementing old ideas.

The ones which seem to be missing from the website (i.e, the ones for which youwould have to buy the book!) include Good Bad Attitude, How to Make Wealth, Mind the Gap, A Plan for Spam, Taste for Makers, Programming LAnguages Explained, The Dream Language.

A couple of these are available from his website as well:

A Plan for Spam
A Taste for Makers

I'd like to now if "The Dream Language" is a genuinly new article, or a rehash of his Being Popular article.

The ones which seem to be missing from the website (i.e, the ones for which youwould have to buy the book!) include Good Bad Attitude, How to Make Wealth, Mind the Gap, A Plan for Spam, Taste for Makers, Programming LAnguages Explained, The Dream Language.

A couple of these are available from his website as well:

A Plan for Spam
A Taste for Makers

I'd like to now if "The Dream Language" is a genuinly new article, or a rehash of his Being Popular article.

The ones which seem to be missing from the website (i.e, the ones for which youwould have to buy the book!) include Good Bad Attitude, How to Make Wealth, Mind the Gap, A Plan for Spam, Taste for Makers, Programming LAnguages Explained, The Dream Language.

A couple of these are available from his website as well:

A Plan for Spam
A Taste for Makers

I'd like to now if "The Dream Language" is a genuinly new article, or a rehash of his Being Popular article.

The ones which are also available on the website are: Why Nerds are Unpopular, Hackers and PAinters, What You Can't Say, The Other Road Ahead, The Hundred YEar Language, BEating the Averages, Revenge of the Nerds and Design and Research.

The ones which seem to be missing from the website (i.e, the ones for which youwould have to buy the book!) include Good Bad Attitude, How to Make Wealth, Mind the Gap, A Plan for Spam, Taste for Makers, Programming LAnguages Explained, The Dream Language.

There are also some on the website which are not in the book.

I had the table of contents from the book and the list of essays from the website reproduced here, but the lameness filter (designed to ensure lameness, I guess) kept saying that the characters per line was 36.

"I suppose I should learn Lisp, but it seems so foreign."

- Paul Graham, Nov 1983

• Paul Graham
  
• Paul Graham: Biography
  
• Paul Graham: A Plan For Spam
  
• Paul Graham: Hackers & Painters
  

• Paul Graham
  
• Paul Graham: Biography
  
• Paul Graham: A Plan For Spam
  
• Paul Graham: Hackers & Painters
  

• Paul Graham
  
• Paul Graham: Biography
  
• Paul Graham: A Plan For Spam
  
• Paul Graham: Hackers & Painters
  

• Paul Graham
  
• Paul Graham: Biography
  
• Paul Graham: A Plan For Spam
  
• Paul Graham: Hackers & Painters
  

And even earlier, as freshmeat points out:

Paul Graham kicked off a flood of mail filters implementing Bayesian filtering with his "A Plan for Spam" article in August 2002, though it was far from a new concept. In fact, ifile has used a Naive Bayes classification algorithm since August 1996 to automatically file mail into folders. In academic circles, Bayesian methods have been used in text classification for many years, and for spam detection prior to Graham, as evidenced by the 1998 workshop paper A Bayesian Approach to Filtering Junk E-Mail by Sahami, et al.

I've never liked the term "computer science." The main reason I don't like it is that there's no such thing. Computer science is a grab bag of tenuously related areas thrown together by an accident of history, like Yugoslavia. At one end you have people who are really mathematicians, but call what they're doing computer science so they can get DARPA grants. In the middle you have people working on something like the natural history of computers-- studying the behavior of algorithms for routing data through networks, for example. And then at the other extreme you have the hackers, who are trying to write interesting software, and for whom computers are just a medium of expression, as concrete is for architects or paint for painters. It's as if mathematicians, physicists, and architects all had to be in the same department.

This has been proposed and implemented several times as an alternative syntax. I think Arc is meant to have this, but it's hardly the first. I don't think it's ever really caught on.

Why?

- Sometimes it's nice to put code into things that don't reliable preserve whitespace, such as, say, comment fields on web sites.

- Parens are well-established in lisp. If you change it you give an additional barrier to people coming from other lisp dialects, without particularly helping people coming from elsewhere.

- Whitespace by itself is not enough. Do you want to write (+ 3 (* 4 5)) across 3-5 lines? Python ends up with fairly complex rules about backslashes, open parens, etc.

- One advantage of lisp is that it's easy to write out from a program. This is really not true of Python.

- If you accept that we need paren syntax, then you can wonder whether indentation should be supported as an alternative. But having two different syntaxes for one language, though an interesting idea, is likely to cause a lot of practical confusion.

So I think all you really want is an editor that ensures the indentation is always valid, and that can highlight parens and do other things. emacs goes a long way, but it could be better -- for example by making outer parens larger, as in TeX-printed or handwritten mathematics.

In my humble opinion what Lisp needs to take from Python is not semantic indentation, but rather a single standard dialect with good OS bindings. The last thing we need is yet another slightly incompatible dialect that can't bind to existing code. Sheesh; I love lisp but lisp implementers really exasperate me.

Why not just use Lisp?

in the last few months the spammers have been intentionally misspelling words in random ways, which reduces the effectiveness considerably. Does anybody know if SpamBayes addresses this issue?

I realize this doesn't quite answer your question, but I suggest you continue to train your filter. Then (like me) you should see the results improve with time.

If you really want to make spam unprofitable, don't prevent people from clicking on the links. Instead, make everyone do it.

Bayesian filters aren't just word filters, in general. Read Paul Graham's article on this. It tokenizes everything, so common features we're not considering (such as bold, red text) also becomes a devistating indicator.

According to the FAQ of SpamBayes (I think), they're always getting suggestions of ways to tweak their algos that would "obviously" improve the result, but in almost every case it either makes no difference or hurts accuracy, when actually tested on real data.

How a programming language bends to someones will is beyond me sorry

FYI Bayesian Filtering isn't quite the same as a Neural network, a notable difference being that with bayes a much greater portion of the behavior learned by the system is easily available for analysis.

Blocklists do not work. Paul Graham's site has done more to reduce my spam burden than all the blocklists in the world combined.

Furthermore, I do not oppose all blocklists. Please don't put words into my mouth like that. In fact the truth is SPEWS is the only major blocklist that I adamantly dislike. I oppose the SPEWS blocklist(s) because the SPEWS blocklists in particular have a false positive rate which is far too high to make them useful except in the most specialized of situations. There are many other blocklists that I do support (e.g. RBL and SBL) because the other blocklists take a more reasoned approach and do more net good than harm.

Sample (for me):

• 1985 * $45k with BS, -$70K going for MS (cost of school + lost wages).
• 1986 * $90k total earned with BS, vs -$140k with MS.
• 1987 * 135k total with BS, vs -$90k with MS.
• 1988 * 180k total with BS, vs -$40k with MS.
• 1989 * 225k total with BS, vs +$10k total earned with MS.

Do you see where this is going? 2027 would be the first year that having an MS would be worth more than having a BS. That is assuming zero cost of money. If the interest rate on student loans was above 0, then it would take longer. One could make a more complicated model with wild assumptions like pay raises, cost of money, inflation, or that you would even be employed. Any engineer or scientist worth their diploma can do the same math. If you had any intention of teaching in High School, an MS would be required. Teaching at the post secondary level would require at least a PhD. Please note that I will make less money in 2004 than I did in 1985.

The leading cause of children entering engineering is to have a family member or friend who is an engineer. As most engineers, scientists and programmers in the US get thrown out of their profession before they finish paying off their student loans, potential students in that career are strongly discouraged from entering the profession. Young adults aren't stupid, if they see a career is a joke, they won't go near it. Gates can whine all he wants, but kids are staying away from engineering and CS with a passion. They can do the math and see that they will never pay off any student loans incurred. Get rid of all the Benedict Arnold CEOs and it may make a difference in a decade.

There was an effort about 10 years ago by Motorola and other electronics firms to get Universities to change engineering curicula to be a 2 year, job preparatory, degree. They did not want to pay for engineers back then, and they still don't. The "book learning" that students learned was not desirable, they wanted training in the specific CAE/CAD software and methodologies used at the specific employer. The half-life of an engineer's career is under 5 years at this time. It would have been far less under the proposed regime: oh, new version of the software came out, you are all fired. Even the H1B abuses that they do is silly when you consider that the CAD/CAE software that engineers use usally has a $20k-$200k annual license fee. They did not want technicians, they wanted engineers but did not want to pay then, nor now for it.

The USA also has a 200+ history of deprecating education, intelligence, intellectuals and "book learning." Things will never get better in the USA unless there is some new sputnik race, and then the change will only be temporary. We never learn, because learning is not cool.

Some interesting essays on Nerd and Uncoolness are here and here. I wish I had seen those 30 years ago when I was about to become a teen, but I also think that I could not have understood them as a teen. Like a lot of things about growing up, they only made sense long after the fact.

Sample (for me):

• 1985 * $45k with BS, -$70K going for MS (cost of school + lost wages).
• 1986 * $90k total earned with BS, vs -$140k with MS.
• 1987 * 135k total with BS, vs -$90k with MS.
• 1988 * 180k total with BS, vs -$40k with MS.
• 1989 * 225k total with BS, vs +$10k total earned with MS.

Do you see where this is going? 2027 would be the first year that having an MS would be worth more than having a BS. That is assuming zero cost of money. If the interest rate on student loans was above 0, then it would take longer. One could make a more complicated model with wild assumptions like pay raises, cost of money, inflation, or that you would even be employed. Any engineer or scientist worth their diploma can do the same math. If you had any intention of teaching in High School, an MS would be required. Teaching at the post secondary level would require at least a PhD. Please note that I will make less money in 2004 than I did in 1985.

The leading cause of children entering engineering is to have a family member or friend who is an engineer. As most engineers, scientists and programmers in the US get thrown out of their profession before they finish paying off their student loans, potential students in that career are strongly discouraged from entering the profession. Young adults aren't stupid, if they see a career is a joke, they won't go near it. Gates can whine all he wants, but kids are staying away from engineering and CS with a passion. They can do the math and see that they will never pay off any student loans incurred. Get rid of all the Benedict Arnold CEOs and it may make a difference in a decade.

There was an effort about 10 years ago by Motorola and other electronics firms to get Universities to change engineering curicula to be a 2 year, job preparatory, degree. They did not want to pay for engineers back then, and they still don't. The "book learning" that students learned was not desirable, they wanted training in the specific CAE/CAD software and methodologies used at the specific employer. The half-life of an engineer's career is under 5 years at this time. It would have been far less under the proposed regime: oh, new version of the software came out, you are all fired. Even the H1B abuses that they do is silly when you consider that the CAD/CAE software that engineers use usally has a $20k-$200k annual license fee. They did not want technicians, they wanted engineers but did not want to pay then, nor now for it.

The USA also has a 200+ history of deprecating education, intelligence, intellectuals and "book learning." Things will never get better in the USA unless there is some new sputnik race, and then the change will only be temporary. We never learn, because learning is not cool.

Some interesting essays on Nerd and Uncoolness are here and here. I wish I had seen those 30 years ago when I was about to become a teen, but I also think that I could not have understood them as a teen. Like a lot of things about growing up, they only made sense long after the fact.

I say, filter everything. Stop the SMTP protocol and hey presto, no spam. Personally, I would make it unlawful to have the 25 port open.

Blacklists like this are the nazi way to fighting spam. Admins (and I'am an admin, too) use their godly powers to crash those spammers -- and just a few nearby unlucky innocent people. I have nothing against personal blacklists, but huge public lists are definitely not the way to go and this is exactly the example why.

Every anti spam tool should be measured in the terms of false positives and false negatives. Well, no false negatives this time, but look at the huge false positive count. And as Paul Graham said:

"False positives are innocent emails that get mistakenly identified as spams. For most users, missing legitimate email is an order of magnitude worse than receiving spam, so a filter that yields false positives is like an acne cure that carries a risk of death to the patient."

I'm an spanish system administrator and I have worked for the biggest ISPs here in the last 10 years, worked for Telefonica too.

1. Telefonica isn't the only one ISP here. Although is the biggest one, the cable provider I work for right now has 500k residential customers. From my head I would say 50% of the market share is on Telefonica hands, but isn't the whole country. Still is a nice way to make up a headline for /. ;)

2. From my experience as ISP sysadmin, I thing blacklisting is a stupid way to fight spam. Is like raiding all the houses of a town because you don't know in house lives the criminal. When you blacklist an entire ISP address space you are blacklisting a 99,9% of ppl who are NOT spammers. Blacklisting advocates would say that if you are a customer of an ISP which doesn't take care of security you should change ISP then. Well, a lot of ppl can't change easily of ISP, I just can't change because the place where I live (rural place outside city bounds) is only served by one ISP.
Changing ISP is a traumatic experience, involving being disconnected while you cease your former ISP service and getting connected again by the new one (in the case of ADSL where the last mile is covered by a single wire which has to change control from one ISP to another), changing email address, changing web hosting, etc, etc... I see changing ISP, like changing phone numbers one of the things I don't want to mess with... and I'm a sysadmin, now think in the average joe who doesn't have a clue of what is happening.

3. 99,9% of the spam I get is from USA space address and is directed at USA ppl. If you take some time to look into what is sold at those mails, most of the times you can only buy it if you live at the states. So I'm amazed that some ppl here are saying "cut all the email from china!", "I never got a real email from a spanish ISP!", if we heard to those idiots we will end talking with our neighbor shouting by the window... I have seen a trend here in /. lately to a strong american point of view, maybe the title should be changed for "news for american nerds" :), still most of the posts fortunately are made with a more (world)wide mind set.

4. I'm totally opposed to make differences between dynamic IP addresses and filtering ports for them. Some ISP gives you static IP address just because is easier for them, so static IP address does not assure you nothing. Also this would make two internets de facto, one some privileges and one without them, and who is going to decide who deserves to belong to each one? Does this means that you aren't allow to have a bussiness with your own hosting with just 1-8 ips? Do I have to ask ARIN or RIPE for a PI space address and run BGP on borders routers just to be qualified to run a mail server?

5. Why nobody is ranting about the old and no up to date SMTP?, no forced authentication, no sender verification, waste of bandwith when attachments are involved... is an old beast which has to die and it's obvious it has a lot of problems addressing the late issues. I would wish all the effort which is put on blacklist would be focused on developing a good standard for mail exchange. SPAM is here to stay and we have to adapt, instead putting stupid patches over old protocols, or thinking about not exchanging mail with other countries.

Look here: Stop spam methods for more up to date methods of fighting spam, still is easier for a dumb sysadmin to just fill in the the form in his server where it says "Put here your blacklist server ip address:"

P.D.: I apologize in advance for my english, think is not my first language. Also my first post in slashdot although I have been reading it for years. :) /me waves

Because Slashdot wasn't when I submitted my site as a newsworthy article some time ago.

In a nutshell, my program, CF13 uses a number of simple, non-mathematic, pattern-matching tests to make it virtually impossible to get English language spam past it. These tests do not require the overhead associated with Bayesian Filtering and its ilk.

I think the key feature to it is to treat as spam all email from unapproved senders that contain more than 'spaces' and alphabetic charaters.

This simple but powerful feature makes it IMPOSSIBLE to conveniently spell email addresses, URLs, postal addresses, prices, and phone numbers. These items are neccessary for e-commerce to take place. Without them, e-commerce is IMPOSSIBLE or at least extremely difficult to conduct. It also treats as spam email containing 'non-ASCII' characters. I have gotten quite a few such emails at another email address I use infrequently--all spam (sales pitches in foreign languages).

As an added benefit, CF13 makes it 100% IMPOSSIBLE to accidentally run malware sent by email provided a particular registry setting has not been compromised. It does this by treating all email and file attachments as 'text files' that can be scanned for malware and handeled safely. Thus, one's PC CANNOT be compromised by a malicious malware HTML webpage or worm/virus/trojan email file attachment.

It also detects 'mailbombing' and handles it a manner that makes it easy to clean up afterwards.

It is probably best to fight spam at the SMTP server level but I have heard it is best to fight spam at the end user level. Both approaches have their advantages and disadvantages so this issue appears to me to be a toss-up for the time being....

I get a ton of spam, check out some of my recent spams and a frequency plot. starting from when I began saving and filtering them. Many thanks to Paul Graham for his plan for spam, or I would be buried by 350 spams per day by now. It is only going to get worse! Based upon how many I get, the probability is more like 95% percent of my email is spam.

The problem is, science *TEACHERS* are not kid-friendly.

And why

CF13 does this by simply comparing all the 'words' in the subject line and body of an email against Grady Ward's Moby single word list and a second, smaller 'spamword' word list derived from the first word list by the user. Both word lists will deem email containing misspelled words or 'spammy' words as spam. Thus....

One more avenue to spam is denied usage by spammers.

By attacking this type of spam technique in this manner, all the overhead associated with Bayesian filtering is 100% completely unecessary.

Lisp should not be used to write a program. Lisp should be used to create a language in which it's simple to write your kinds of programs, even if they need unusual control structures such as Prolog backtracking. There are excellent examples in Paul Graham's On Lisp.