Slashdot Mirror

I recently read a PhD dissertation discussing (among others) a java-based grid environment. At first the entire idea sounded odd to me because of an expected performance drop. However, cases where java can be as fast as and sometimes even faster than optimized C have been shown before.

In the dissertation, two papers are mentioned that compare java to C and apparently show positive results for java. I haven't actually read these papers, by the way. However, their titles suggest that they might be on-topic here.

The first is "Benchmarking Java against C and Fortran for Scientific Applications" by J. M. Bull, L. A. Smith, L. Pottage and R. Freeman

Second, and judging from its title perhaps less on-topic is "High-Performance Java Codes for Computational Fluid Dynamics" by Christopher Riley, Siddhartha Chatterjee and Rupak Biswas

finally, the dissertation can be found on Rob van Nieuwpoort's publication list and is titled "Efficient Java-Centric Grid-Computing"

As citeseer is always crawling, even without a link on slashdot, I was afraid to post this... but then I remembered that noone here actually reads the FM :) Jokes aside, it is a great resource and I hope that they get some mirrors soon.

I recently read a PhD dissertation discussing (among others) a java-based grid environment. At first the entire idea sounded odd to me because of an expected performance drop. However, cases where java can be as fast as and sometimes even faster than optimized C have been shown before.

In the dissertation, two papers are mentioned that compare java to C and apparently show positive results for java. I haven't actually read these papers, by the way. However, their titles suggest that they might be on-topic here.

The first is "Benchmarking Java against C and Fortran for Scientific Applications" by J. M. Bull, L. A. Smith, L. Pottage and R. Freeman

Second, and judging from its title perhaps less on-topic is "High-Performance Java Codes for Computational Fluid Dynamics" by Christopher Riley, Siddhartha Chatterjee and Rupak Biswas

finally, the dissertation can be found on Rob van Nieuwpoort's publication list and is titled "Efficient Java-Centric Grid-Computing"

As citeseer is always crawling, even without a link on slashdot, I was afraid to post this... but then I remembered that noone here actually reads the FM :) Jokes aside, it is a great resource and I hope that they get some mirrors soon.

What I recently discovered was that this form of autobiographical 'drivel' is by no means a new form of literary expression.

The term I-novel started to be used in literary circles (called bundan) around the mid-1920s. Originally it meant contemporary autobiographical sketches whose authors appeared to write directly about their personal lives for a closed circle of fellow writers.

Er, well, the building basically exists, but the students in your story do not.

The building is Kresge Auditorium. It was designed by Eero Saarinen, one of the most famous architects of the 20th century. He also designed (e.g.) the St. Louis Arch and the TWA Terminal at JFK Airport.

A somewhat biased but detailed view of Kresge Auditorium is available here. As you can see, no PhD theses are mentioned.

The building's roof is a single thin concrete shell. The original design was very ambitious, such that the roof was to be supported only at the three points where the shell contacts the ground. The design was later changed so that the mullions in the large banks of windows would bear some of the load.

Kresge Auditorium was one of the first buildings of this type. More thin shell concrete structures available here.

It ain't just professors, or those in acadamia; part of the salary review process, at most private research institutes, includes an accounting of the number of peer-reviewed external publications which an individual has to their name, in the preceeding fiscal year.
It adds prestige to the corporation.

It's the same, to a large extent, with government research groups (i.e. NIST, &al) -- although there, you're gaining this coin of prestige for yourself, and your department (thus attracting more post-docs to do more work for you...).

It's not so much an issue of if people are paying to read your reports, though -- your company/school/department probably doesn't see any direct fiscal return on that... So... so long as there were a valid way to measure the prestige of the publication (perhapse something a'la CiteSeer, or some-such... (while maintaining a verifiable, and accountable peer-review process) and convince the bureaucratic structure to adopt that metric then, I suspect, the greatest hurdles would be taken care of.

For whatever reason, no one has mentioned the very important link between grant money and publications. Its not as if people are submitting publications for shits and giggles. If you use grant money to publish, especially in more major journals in your field, the likelihood of having grant renewed/accepted increases substantially. It is this grant money that gets used for subscriptions, academic fees which pay for subscriptions, etc.

I think that no one would argue that refereed publications are a necessity. Even if it doesn't work, its better than an open system such as slashdot, where the only qualification is the time spent on-line, but the purpose, is of course, different. The real crime is that the middle-men involved, kluwer, elvesier, etc., make a significant amount of money off of these publications, but are far worse, on-line than publications such as citeseer, which would be perfect (free and easy) if it allowed a distinction for refereed publications.

I, personally, see print journals becoming a thing of the past, as they are non-searchable and slow to access (i.e., going to the library or requesting an item). Additionally, the charge for these articles, individually, is outrageous, $15-35? (thank you IEEE) How about $5?

Sometimes papers are submitted to journals, and are hard to find elsewhere. Most of the time, an e-mail to the author will get a response, or it can be found using a search engine.

It's been a long time since I have looked in a paper journal, yet I still know of universities who shun electronic access...

D'oh! Arizona, not New Mexico. Near Flagstaff.

I got too curious and looked up the reference myself... p. 155 of The Indispensable Calvin and Hobbes.

It really shows the McAfee guys for what they are: a bunch greedy ignoramuses who want to make money with other people's ideas and can't even do their homework.

The same is true for the USPTO except they are lazy instead of greedy.

-- Anonycous Moward

There are serious investigations into making cache optimized algorithms. For example, the matrix transposition and array index bit reversal algorithms have been investigated in two papers. Also, Bailey's 4-step and 6-step FFT algorithms are also cache efficient. The latter example shows that a complex algorithm such as a FFT can be made cache efficient with the sacrafice of only a few extra computations. Perhaps it would be prudent to use a hybrid ray-tracer/polynomial renderer to section each portion of the screen into regions that will only access a particular portion of memory. In fact, texture mapping is a lot like that. But I propose that we section the geometry into sections that are localized in memory. This will require more computation in the form of checking which ray goes where but it might be possible to create a viable ray tracer/polygon renderer that produces images of ray tracing caliber. By polygon renderer I mean the renderers that we currently use in gaming.

Some references about cache efficiency.

There are serious investigations into making cache optimized algorithms. For example, the matrix transposition and array index bit reversal algorithms have been investigated in two papers. Also, Bailey's 4-step and 6-step FFT algorithms are also cache efficient. The latter example shows that a complex algorithm such as a FFT can be made cache efficient with the sacrafice of only a few extra computations. Perhaps it would be prudent to use a hybrid ray-tracer/polynomial renderer to section each portion of the screen into regions that will only access a particular portion of memory. In fact, texture mapping is a lot like that. But I propose that we section the geometry into sections that are localized in memory. This will require more computation in the form of checking which ray goes where but it might be possible to create a viable ray tracer/polygon renderer that produces images of ray tracing caliber. By polygon renderer I mean the renderers that we currently use in gaming.

Some references about cache efficiency.

A small snippet from the slackware-current ChangeLog.txt:

"I also noticed that the ATI Radeon binary drivers designed for XFree86 4.3.0 do not work with XFree86 4.4.0, but do work with the X.Org release."

Full ChangeLog.txt is available here, for those who are interested. The XFree86 to X.Org change is toward the bottom of the entry for Sun May 30 01:06:39 PDT 2004.

Wrong. Good programmers understand how their environment treats numerical values. In Perl(and several other languages) 3 and 3.0 are treated identically. There is nothing wrong with comparing them. Also, based on your comments in this thread it seems clear that you don't know what you are talking about. Go read this, and the other reading material referenced here. If you don't want to, or can't, read and understand that material then please quit spouting off and spreading your misconceptions to others.

Look, before you say anything else, MattRog, I think you ought to read this

Not only does this relay the well-known problem of finding a subtree by reachability (and that's just with the with homogeneous trees you can represent... still busy?) for the relational calculus, but has a formal proof that the extra (second order) features of SQL do not give it this power either... Step up SQL3 (by no means whatsoever faithful to the relational calculus)!

Regarding the quotes: fair enough. Quoting someone else's opinion on a subject is not proof of anything. We could both come up with quotes all day long that support our respective positions.

Regarding the primary source of your information: yes, I am familiar with the Talk Origins archive; your first two citations came directly from the same page, and the second two involve topics from that same page (I didn't look at all the bacterial flagellum articles to find the other specific articles you cited).

...the appearance of a totally new metabolic pathway for sugar metabolism in the genus Klebsiella is misleading; the pathway was not totally new and no mutation was shown to have occurred. At no time did the researcher point to a new gene and say "this was not here before." A previously unobserved trait was expressed and enhanced by "unnatural selection" in the artificial environment but in no way did the study show that the "new" ability to metabolize D-arabinose was the result of a beneficial mutation. It is against all reason that the one "beneficial mutation" that the researcher was looking for just happened by "faulty repair of genetic damage" or "copying errors," especially since bateria do not reproduce sexually.

...There are plenty of intermediate forms preserved in the fossil record. To name but a few; there are some beautifully preserved fossils of ammonites that show a progression of structures (the curved shells unwinding over millions of years), there are clear stages preserved in the development of flowers, and best of all, there is the evolution of mankind over the past few millions years - plenty of clear intermediate stages. A good website for the convinced skeptic is http://home.entouch.net/dmd/transit.htm, which shows the transition from fish to amphibian.

Wrong. it shows a creature that has some morphological similarities to fish and amphibians, quite a large difference. Taking a few similar forms and artificially arranging them to look like a transition does not mean they are transitional forms. The "progression of forms" you mention has another small problem; the "progression" is in many cases out of order in the fossil record. And the "transitional forms" of man were in many cases contemporaries of each other. The whole idea that organisms that are morphologically similar are therefore related is misleading in any case; is a platypus a transitional form between reptiles and mammals because it shares some morphological features with both? Of course not, and no biologist will tell you otherwise. Similar morphology does not indicate that any two species are related. As an example, a recent analysis of the genetic structure of different species of water fowl has revealed that the flamingo, while morphologically similar to the heron, crane, and spoonbill, is actually most closely related to the grebe (a small duck) than it is to any of the other wading birds.

By the way; the title of the page you referenced is "TRANITIONAL FORMS." Makes one call into question the thoroughness of the author, don't you think?

Then you say...There is no such thing as a 'nascent' organ in 'pre-functional form', and there never has been. At each stage, the change has to be of increased benefit to the organism. Evolution does not have the ability to look into the future and plan the construction of an organ.

I agree. However, the same page you cite categorizes the fins of Acanthostega as "half-evolved legs;" this shows his misunderstanding of the very theory he is espousing; as though evolution had legs in mind but was only halfway there. As for the heat-sensing pits of vipers; talk about a straw man! "Maybe" "someday" it "could" evolve into an eye; how is this a tr

Um, no. First of all, the jury selection process has become the jury tampering process. A jury of peers should be a randomly-selected group of eligible people, but it's more or less handpicked nowadays. The verdict is often decided by which lawyer is craftiest in "disqualifying" potential jurors. Jurors in the pool should not be asked any questions aside from:

Do you personally know the plaintiff or defendant?

Do you have any hearsay knowledge of this case?

Have you or any member of your immediate family ever been the victim of a similar crime?

Anything else is jury tampering, and jurors should refuse to answer!

Second, there is the question of jury nullification. Judges and prosecutors seldom inform juries of their right and responsibility to return a "not guilty" verdict if they feel that the law does not reflect the values of the community or has not been applied appropriately. Jury nullification is the final check against the legislative and executive functions of government, and it has a long and established history. However, citizens have been harassed and even charged with contempt of court for exercising this sacred right.

Finally, note that on modern drives, you can seek all the way across the disk in only about 30-50% more time than you can seek a short seek. Thus keeping your blocks close to the current cylinder but not in it has very limited value. Note that this is not the case on optical (CD/DVD) disks.

Bzzt. Wrong. Thanks for playing.

Seagate Barracuda 7200.7 Plus product manual. This is a modern drive. Look at page 17 of the PDF, section 2.7, "Seek Time". Track to track seek is listed as sub 1ms, while average seek is 8.5ms (for read). Latency (the rotational aspect), which is separate, is an average of 4.16 (as it should be for a 7200 RPM drive). So a short seek is 8 times faster than an average seek, much less a whole-drive seek. So keeping your blocks close to the current cylinder but not in it still has high value.

Additionally, if you can keep your data in the same track, you don't have any seek time, just rotational latency. And the size of track groups has been growing as densities have been growing. So there are lots and lots of blocks in the same track that aren't within readahead range.

And fragmentation is not a bugaboo. It's a fact. When you have random allocation on a volume, it will get fragmented.

Now you misuse the term "random allocation". When talking about disk, random allocation means that you randomly choose your next block--which certainly will cause fragmentation. I think what you are looking for is random file creation. However, under FFS, if you maintain sufficient free space, it is very unlikely that real fragmentation will occur even with random file creation/deletion. Yes, you won't be able to store all of your files contiguously. But as the grandparent points out, the old FFS block allocator finds "nearly optimal" blocks. From the original paper on FFS, you don't get serious fragmentation-related performance issues until you reach 90% disk utilization (there's an '86 FFS paper I can't find an electronic copy of which does a better analysis, but even the '84 paper has the 90% figure). At 90% utilization, nearly every file system ever starts getting severe performance problems due to fragmentation.

You can go back after the fact and unfragment it, but doing so in any serious fashion when writing files actually degrades performance due to the extra effort required.

You should read about log-structured file systems sometime. Like Sprite. The base idea of a log-structured file system (LFS) is that you don't try to keep your blocks the same. You write a log of changes, and stream that out continuously to make maximum use of your available write bandwidth. This has the severe downside of causing horrific freespace fragmentation, since every change to a file means the affected blocks are reallocated, and the old blocks are now garbage. So you have continuous freespace compaction (called segment cleaning in the paper). LFSs didn't catch on because at the time, they needed a large amount of free space compared to more traditional "overwrite" file systems to maintain performance. However, recent work with log file systems shows that the changing performance characteristics of disks (much higher bandwidth but same seek/rotational delays) have tipped the balance to log structuring. Regardless, the cleaning process is precisely going back and unfragmenting your data, and is necessary for a LFS.

HFS+ does nothing to prevent fragmentation except for use super-clusters.

As pointed out by others, Apple's later implementations do defragmentation of smaller files when they are accessed.

NTFS could do this, but I believe they do not. However, NTFS on servers has an allocation block size of 8K

Apple isn't patenting "translucent windows", they are patenting a specific method for choosing to make windows translucent. The method seems pretty hokey to me (windows automatically become more translucent over time and eventually let events pass through), and probably has horrible usability problems. I can just see the support calls: "but my Microsoft Word window was there a few minutes ago, and after I came back from getting a cup of coffee it was just gone".

In this wonderful world of software patents, the patent may be valid, but it is not relevant to anything real.

If you want to read about good uses of translucency in user interfaces, see this survey from 1994 (long before OS X).

Since cattle need to drink water, they add to the load on fresh water demands. Cattle consume from 1 (for a 1 month old) to many gallons (for a lactating cow) per day. Ref: Water intake and quality for cattle

You should try citeseer or one of its mirror too.
Most paper are free for download.
IEEE is another website or ACM queue.

http://citeseer.ist.psu.edu/cis?cs=1&q=stack+based +virtual+machine

You can do this today with Chromium.

Chromium replaces your OpenGL library with one that farms the OpenGL drawing out to multiple machines. It's how display walls are built.

You can use the same technique for multiple card in the same box.

I have no first hand knowledge in the subject, but perhaps this link to citeseer will help.
-blue

I have no first hand knowledge in the subject, but perhaps this link to citeseer will help.
-blue

Seriously, that's old. I'm a computer vision Ph.D. student, and there now are much faster methods. I'll just refer to my old comment.

A demo can be found here. You can contact me for more details...

Current really fast methods use cascades of very simple classifier that are very weak themselves, but very strong when combined. The work of Viola & Jones is what most of the stuff is centered around nowadays.

Do your own here:

http://argus.cs.unimaas.nl/fddemo

You don't need to hear individual bits. You just need to see a statistical difference in speed over iterated modular exponentiations. See Kocher's seminal 1996 paper on timing attacks for a good explanation of this type of analysis. Note that Kocher is measuring operation times on the order of 10ms. This is well within-bounds for an audio measurement.

Not being a professional cryptographer myself, I'm really not qualified to say for sure how practical this attack is. However, I can't rule it out on the basis of my amateur knowledge. I'd say there's a good chance it could work.

Indeed, the SVM can be used to model a linear decision boundary (or, alternatively, do regression) in any feature space. The kernel has to comply to Mercer's theorem for most kernel machines, but not for all (e.g. not for the relevance vector machine).

Later addition? Nonlinear kernels were already used even before the SVM was called the SVM. See here. Perhaps you refer to all tutorials, which make it look like it was a later addition.

>What is so hard about using "em" as the base measurement for the layout? Also, the main content text size should always be 1em, i.e. not specified, allowing the browser's default size to be used.

• Use a browser safe color palette (216 colors).
• Use contrast. Dark text on pale background is better on a computer than light text on dark.
• Avoid vibrating contrasts, especially red and green next to each other (color blindness) .
• Use bright or different colors for emphasis.
• Use the 3-Color Rule.
• Never use colors from the opposite ends of the color spectrum together.

see this link to ACID whitepaper

also see this portion of the Inferno manual.

and this entry on citeseer.

*the* AC

The question is why didn't the design hubble to be repaired in this way in the first place? The cost launching the space shuttle is around 375 million dollars.. Probably more for a space walk..

I don't accept that you can't design a repair bot for under that launch cost?

Simon

A friend of mine did his thesis on a Garbage Collector that is parallell, concurrent, real-time and (mostly) tagless. We have it running in a real compiler. (Here's a later paper he wrote.) So real-time GC is absolutely possible.

How about one of the earlier comments to the effect that mark-and-sweep type algorithms page-faults all the memory used by an application?

Not all garbage collectors are mark-and-sweep. Also, copying GCs have much better fragmentation than malloc, so if you're concerned about keeping the working set small, a copying GC is absolutely the way to go.

DPGA-Coupled Microprocessors: Commodity ICs for the Early 21st Century (1994)
Reconfigurable Architectures for General-Purpose Computing (1996)
Transit Note 118 Notes on Coupling Processors with Reconfigurable Logic

DPGA-Coupled Microprocessors: Commodity ICs for the Early 21st Century (1994)
Reconfigurable Architectures for General-Purpose Computing (1996)
Transit Note 118 Notes on Coupling Processors with Reconfigurable Logic

DPGA-Coupled Microprocessors: Commodity ICs for the Early 21st Century (1994)
Reconfigurable Architectures for General-Purpose Computing (1996)
Transit Note 118 Notes on Coupling Processors with Reconfigurable Logic

I've just gone through the same process of learning LaTeX. However, I'm an OS X user and I found Mac-Tex at Penn State to be a very good resource. I chose TexShop for my front end iInstaller to install the LaTeX backend. You can also use Fink to install your backend but I didn't feel like comand line install this time as suggested previously.

Other than getting the software installed, I simply used Google for tutorials on LaTeX and BibTex.

> This will provide material for a substantial test of Bill Dembski's theories about the limitations of evolutionary algorithms.

The "theory", which Dembski gratuitously mis-applies, is Wolpert & McReady's No Free Lunch Theorem.

Dembski is nothing but a creationist apologist, relying on pseudo-science and obfuscation to give creationism a glamor of scientific respectability among the ill-informed.

'cwm' isn't really an exception, is it's a borrowed word from Welsh, where w is a vowel. See also this interesting mailing list post.

At my school my Physics 211 (Mechanics) professor adds a semgent to almost every leture called "Hollywood Physics." It similar to the idea of the article. His aim is to bascially prove, and show why a particular stunt or action scene would work, or disprove the ones that would not work. He formulates some assumptions and then solves the problem with formulas we are using during that chapter. It's one of the more interesting parts of the physics lectures, that is, the ones I can crawl out of bed for. It sure beats massless pigs on a frictionless inclined plane.

At my school my Physics 211 (Mechanics) professor adds a semgent to almost every leture called "Hollywood Physics." It similar to the idea of the article. His aim is to bascially prove, and show why a particular stunt or action scene would work, or disprove the ones that would not work. He formulates some assumptions and then solves the problem with formulas we are using during that chapter. It's one of the more interesting parts of the physics lectures, that is, the ones I can crawl out of bed for. It sure beats massless pigs on a frictionless inclined plane.

I take serious issue with these institutions (namely Penn State) that take student money and divert it to services that students have not authorized the institution to do so. What if a student doesn't even own a computer? Should that student still have to pay for the service through their tuition?

I've just accepted an offer for a little more than 50K a year (before taxes), in a suburb of a medium-sized city where cost of living should be relatively low. The offer is from a very large company. I'll be graduating in May with a degree in Computer Science from psu.

"How are you going to adjust your encryption when quantum computers will make most encryption schemes obsolete? "

Why FUD my friend? This just isn't true!

The truth about quantum cryptography is that RSA and DH will be destroyed by quantum cryptography. This is due to the work of Shor who famously proved that you could factor in cubic time.

This sounds bad but we've already had good success in performing quantum key exchanges (that are unbreakable in a theoretical sense).

What does this mean for symmetric cryptography such as AES? Well, Quantum Computers that deploy Grover's algorithm, can search unordered lists in under sqrt(n) operations. A normal computer does this in an average of n/2 steps. The key space of a cipher is an unordered list so we'd only have to double our keysizes to avoid the "Grover attack".

Clearly more research is needed but the quantum future is bright as far as cryptography is concerned.

Simon.

Peercy and Olano (Click on "PDF" in the upper right)

Presentation

ASHLI

GPGPU

More than Moore's Law

Moore's law : still for wimps

Using programmable graphics hardware (possibly through OpenGL) for final rendering is not that far off. (Definitely not in real-time, but as a more cost-effective way to do it, anyway.) Especially with the massive parallelism of rendering, and the fact that GPUs are far outpacing CPUs in terms of their speed and transistor counts.

OpenGL is much more similar to micropolygon rendering (REYES) than it is to raytracing in the first place. The shaders are where you spend all of your time, anyway.

Heck, do you think nVIDIA bought ExLuna (Larry Gritz, author of BMRT, and former Pixar employee) just for the fun of it?

Software for translating from RenderMan Shading Language to Cg?

And what about RenderMonkey supporting RenderMan?

Do you even remember PixelFlow from Pixar? Do you see the name Marc Olano on that paper? The same Marc Olano who talks about rendering on consumer-level graphics hardware? These things have far more in common than you seem to realize.

The charged particles causing soft errors can come from three different sources.

-There are particles originating from radioactive impurities in the semiconductor material itself and from surrounding packaging and contact materials such as lead (Pb). Minute levels of such impurities suffice to create problems. These impurities emit alpha particles.

-There are low-energy neutron particles coming from space that are interacting with a certain type of boron (B10) present in the semiconductor material like BPSG. They also create energetic alpha particles through a nuclear reaction. Although they clearly originate from the outside, they only create problems if B10 is present.

-Finally, there are high-energy neutrons coming from space. These effects are very pronounced at higher altitudes and have been known for years to cause problems in space applications but even in commercial jetliners although less of course.

Well, he certainly did a pretty decent job analyzing the 1988 Internet Worm.

Anticipatory scheduling:
A disk scheduling framework to overcome deceptive idleness in synchronous I/O (2001)

Sitaram Iyer, Peter Druschel
18th ACM Symposium on Operating Systems Principles

ACM portal

Using the old citeseer trick, you can read the PDF version here:
Citeseer paper version
PDF version

Don't SLASHDOT citeseer!
There is more than one citeseer mirrors, use google:
Google Citeseer paper search

Enjoy!

Anticipatory scheduling:
A disk scheduling framework to overcome deceptive idleness in synchronous I/O (2001)

Sitaram Iyer, Peter Druschel
18th ACM Symposium on Operating Systems Principles

ACM portal

Using the old citeseer trick, you can read the PDF version here:
Citeseer paper version
PDF version

Don't SLASHDOT citeseer!
There is more than one citeseer mirrors, use google:
Google Citeseer paper search

Enjoy!

Only if Homer does the voice...

There was an video segment on Discovery science about the evolution of insect wings. The basis of the research was based on stonefies and demonstrated how gills could have evolved into stubs which would not have given flight by themselves, but would have provided enough acceleration for the critter to escape from predators. The arms race between these two species would have forced ever faster acceleration to occur, until being fast enough to make flight possible.

This is starting to change in computer science, although other fields are a long way behind. I'm studying for a PhD at the moment, and most of the papers I need are available online, either on the authors' websites or on Citeseer. Even in CS, older papers are less likely to be available, but most of the work in my area was published in the last four years or is still awaiting publication. That's the other advantage of publishing online - the process of getting a paper reviewed and published can take years, so in fast-moving fields the journals are really an archive of significant work rather than a news medium. To keep up with recent work you have to look online.

Of course, the problem with self-publication is lack of peer review. However, Citeseer does a pretty good job of finding significant papers based on the number of citations (think Pagerank), and the database of citations also helps you to find papers that might contradict or reinforce the conclusions of the paper you've just read. This makes it less important to have editors filtering out biased or unreproducible results.

I hope that authors in other fields will start to embrace online self-publication. Unfortunately, many institutions see publication count as a good measure of an academic's standing, partly because the peer review process tends to ensure that a frequently-published author is well respected in his or her field. If insitutions started to pay attention to citation count instead, self-publication would become a viable alternative to journal publication, saving students and taxpayers an awful lot of money.

here fix your link
link is here you had a space in "spring"