Sure, you are still out of 100K, but if you could afford (say) a 500K house, you most likely can afford again a mortgage for 100K.
Earthquake insurance is expensive, and this is
why it is not incredibly foolish not to have one.
If you bought the house, you can usually also afford to rebuild.
On the other hand, if you lost most of the value of the property,
so that nobody would lend you money to rebuild,
you would be in far more trouble.
Weird, but true. There is an advantage in paying a lot for the land: it's harder to destroy than the house!
What is expensive is the land, not the building per se. If my house was razed by fire or earthquake, I would lose at most 30% of what it costed me - the rest is the price of the land.
This also means that banks would have no problems using the land as collateral to finance my rebuilding the house.
So you see, high house prices are actually a kind of self-insurance!
I am speaking from the SF Bay area.
For computer science, I often have much more
luck using Google Scholar than Inspec.
Google Schoolar is superior especially when you
don't happen to know exactly the best search
terms for what you are looking for; Inspec,
even with "search in the abstract" enabled,
does a much poorer job. I guess this is because
Inspec does not search in the body of the document, unlike Google.
The only problem with Google scholar is that it
points to pay-for-view sites for the articles
(such as the ACM, IEEE web sites) even when the
authors have posted a copy of the paper on their
web pages. But another quick Google (non-Scholar) search and the problem is solved.
And anyway, as I work in a university who has access to all kind of sites, this is
not really an issue for me.
I think the social stratification is definitely happening. But it's not primarily a matter of money: many times, you can go as far as a PhD and be supported throughout the studies by grants and fellowships (in fact, in CS, being supported financially by grants during a PhD is the norm rather than the exception).
It is also a matter of values.
For the gifted, the choice is primarily a matter of valuing education and long-term prospects more than immediate earnings.
Even if you can get an education w/o having to pay for it (if you are good), you need to make the decision to accept (relative) poverty while studying, rather than getting a job right away and make more money in the short term.
Of course the relative poverty of students is mitigated by the fact that being at a University is mostly a great experience.
However, you are delaying things like a real salary, being able to afford a half-decent car, living in your own apartment, making retirement contributions, etc etc, to 4-8 years later.
I admit it's easier to decide to embark on a university education if you family can financially back you during your studies, so that you don't quite have to limit your expenses so much.
I also admit that if you are wealthy, you can
buy yourself an education within limits
(beyond the MS level, nobody cares if you can pay,
at least in CS/Engineering, but everybody cares
if you are good).
But I believe it is also a matter of relative importance, immediate money vs. long-term career and education, and I believe that this is part of a scale of values that gets passed from parents to children along families, and represents a form of "class".
So the children of low-income parents that keep studying until they get advanced degrees were already part of the class that values education more than immediate money.
I write research papers. I keep each paper
in a directory, and I keep the paper directories
organized by year (it's funny, but when I look
for some old paper I wrote, I remember better
the year and the place of publication than the
title).
If I look for the introduction of a paper published in XYZ 2000, I know it is in a file
most likely called intro.tex, included from a
file called paper.tex (or perhaps submission.tex), and that in somewhere in a file
paper-web.tex is written "In the Proc. of XYZ 2000". Would I have to attach metainfo to all the files for a paper (typically, 10-30, including figures etc) to be able to relate them? It seems to be a more boring way that just putting them into the same directory. And I would not be able to do "ls". And, at the last count I have 244 files called "intro*.tex" on my PC (no, I did not write all of them), and I may have multiple copies of the intro for the same paper (version 1, submitted version, longer version,...). How am I supposed to find those? Directories organize not only the files, but your mind as well!
But sure, progress will be when instead of doing
cd papers/05/XY
ls
emacs intro.tex
I will have to tag files, enter searches, and
click with the mouse 20+ times before I can finally edit my introduction.
No no, listen to rar, he has it right.
I also use my PCs to get the job done,
but that requires unfortunately package
upgrades. For example, I need the latest
openoffice to open and edit better the MS
documents I get emailed, I need the latest
cinepaint to do a bit more with image manipulation,
I need good USB support (Woody does not have it)
to read memory sticks, and so on and so forth.
You cannot reasonably expect desktop users
that just need to "get their job done" to use
a 3-year old distro, because they will want to
take advantage of new perhipeals, and new
features, that will enhance their productivity
in getting the job done.
It is a real pity that testing does not get
security upgrades; it means essentially that
Debian has a good server distribution (the stable one) and a good distribution for experimenters (the unstable one), but no distribution aimed at desktop users.
No wonder Ubuntu and other distros are
becoming popular.
I am VERY impressed by the photos - certainly
better, artistically, of what Reuters/ANSA
or the like usually produces!
As for the dead cows, I bet they ate grass
contaminated with rocket fuel - it can be
very poisonous. I am not sure what they
use, but hydrazine, for instance, is
very poisonous.
Thanks - you are right.
So now I have to find a way to compile that library... there are many machines that need cvs/svn
but not libapr. But all in all, it might be too much hassle for me to install libapr on all machines (including cygwin ones) just to use svn.
Another reason not to use svn is that it
doesn't work on old versions of linux.
For instance, if you try to compile it for
Debian Woody (stable), it compiles fine,
but then it gives a library error when started.
Same thing on Redhat 7.3.
Who cares, you will say?
Well, I use Debian stable for all my critical
servers, due to the ease of doing security
updates. Moreover, many of the people that
need access to my cvs/svn server are running old versions of linux.
So, unless Tigris fixes this library problem,
I will stay away from svn.
Yes, I also get several spams a month from this group of people organizing conferences in Orlando, Florida. They ask me to "submit a paper, or if I like, organize an invited session" - so that I invite my friends and make the conference larger.
Verbatim from one of their emails:
We would also like to invite you to consider the possibility of organizing
an invited session in the area or topic of your research interest or in the
context of your experience. To do so fill, please, the web page form given
at (http://www.cyberinformatics.org/rmci05/reviewers/ register_reviewer.asp)."
My guess is that the organizers make money out of the conferences, and people who want to buff their resumes submit to it. I do not believe papers are reviewed in any way (that would be work!).
This should not at all be construed as an assessment of the general quality of conferences. These days, the review process in prestigious conferences is usually better than in top journals.
For prestigious conferences, it is a honor to be on the program committee, so top people accept to do this. This is different from journal reviews, that are a lot of work, and generally little honor (because the names of the reviewers cannot be publicized in the same way that program committee membership is). Furthermore, journal papers generally are reviewed by 2-3 people, whereas 4 reviews is the norm for top-quality conferences. At least, this is the situation in computer science these days.
Of course, as this example shows, one needs to know which conferences are serious and which ones are not.
Am I the only one that finds it very disturbing the fact that, in order to search your PC, you actually send out the search query?
So Google can build for free an incredible amount of statistical knowledge on what users have on their PCs? Just how much is that worth? And, is the information encrypted? For otherwise, it would certainly be of interest to know what e.g. Intel's engineers are searching on their PCs...
It amazes me how much information people are willing to give out for free in exchange of a little convenience.
Slashdot Mirror
Earthquake insurance is expensive, and this is why it is not incredibly foolish not to have one. If you bought the house, you can usually also afford to rebuild. On the other hand, if you lost most of the value of the property, so that nobody would lend you money to rebuild, you would be in far more trouble.
Weird, but true. There is an advantage in paying a lot for the land: it's harder to destroy than the house!
What is expensive is the land, not the building per se. If my house was razed by fire or earthquake, I would lose at most 30% of what it costed me - the rest is the price of the land. This also means that banks would have no problems using the land as collateral to finance my rebuilding the house.
So you see, high house prices are actually a kind of self-insurance! I am speaking from the SF Bay area.
The only problem with Google scholar is that it points to pay-for-view sites for the articles (such as the ACM, IEEE web sites) even when the authors have posted a copy of the paper on their web pages. But another quick Google (non-Scholar) search and the problem is solved. And anyway, as I work in a university who has access to all kind of sites, this is not really an issue for me.
Thumbs up to Google Scholar!
For the gifted, the choice is primarily a matter of valuing education and long-term prospects more than immediate earnings. Even if you can get an education w/o having to pay for it (if you are good), you need to make the decision to accept (relative) poverty while studying, rather than getting a job right away and make more money in the short term. Of course the relative poverty of students is mitigated by the fact that being at a University is mostly a great experience. However, you are delaying things like a real salary, being able to afford a half-decent car, living in your own apartment, making retirement contributions, etc etc, to 4-8 years later.
I admit it's easier to decide to embark on a university education if you family can financially back you during your studies, so that you don't quite have to limit your expenses so much. I also admit that if you are wealthy, you can buy yourself an education within limits (beyond the MS level, nobody cares if you can pay, at least in CS/Engineering, but everybody cares if you are good). But I believe it is also a matter of relative importance, immediate money vs. long-term career and education, and I believe that this is part of a scale of values that gets passed from parents to children along families, and represents a form of "class".
So the children of low-income parents that keep studying until they get advanced degrees were already part of the class that values education more than immediate money.
I write research papers. I keep each paper in a directory, and I keep the paper directories organized by year (it's funny, but when I look for some old paper I wrote, I remember better the year and the place of publication than the title). If I look for the introduction of a paper published in XYZ 2000, I know it is in a file most likely called intro.tex, included from a file called paper.tex (or perhaps submission.tex), and that in somewhere in a file paper-web.tex is written "In the Proc. of XYZ 2000". Would I have to attach metainfo to all the files for a paper (typically, 10-30, including figures etc) to be able to relate them? It seems to be a more boring way that just putting them into the same directory. And I would not be able to do "ls". And, at the last count I have 244 files called "intro*.tex" on my PC (no, I did not write all of them), and I may have multiple copies of the intro for the same paper (version 1, submitted version, longer version, ...). How am I supposed to find those? Directories organize not only the files, but your mind as well!
But sure, progress will be when instead of doing
cd papers/05/XY
ls
emacs intro.tex
I will have to tag files, enter searches, and click with the mouse 20+ times before I can finally edit my introduction.
It is a real pity that testing does not get security upgrades; it means essentially that Debian has a good server distribution (the stable one) and a good distribution for experimenters (the unstable one), but no distribution aimed at desktop users. No wonder Ubuntu and other distros are becoming popular.
As for the dead cows, I bet they ate grass contaminated with rocket fuel - it can be very poisonous. I am not sure what they use, but hydrazine, for instance, is very poisonous.
Bah, perhaps I am turning into a luddite.
Who cares, you will say? Well, I use Debian stable for all my critical servers, due to the ease of doing security updates. Moreover, many of the people that need access to my cvs/svn server are running old versions of linux.
So, unless Tigris fixes this library problem, I will stay away from svn.
Verbatim from one of their emails:
We would also like to invite you to consider the possibility of organizing an invited session in the area or topic of your research interest or in the context of your experience. To do so fill, please, the web page form given at (http://www.cyberinformatics.org/rmci05/reviewers/ register_reviewer.asp)."
My guess is that the organizers make money out of the conferences, and people who want to buff their resumes submit to it. I do not believe papers are reviewed in any way (that would be work!).
This should not at all be construed as an assessment of the general quality of conferences. These days, the review process in prestigious conferences is usually better than in top journals. For prestigious conferences, it is a honor to be on the program committee, so top people accept to do this. This is different from journal reviews, that are a lot of work, and generally little honor (because the names of the reviewers cannot be publicized in the same way that program committee membership is). Furthermore, journal papers generally are reviewed by 2-3 people, whereas 4 reviews is the norm for top-quality conferences. At least, this is the situation in computer science these days.
Of course, as this example shows, one needs to know which conferences are serious and which ones are not.
It amazes me how much information people are willing to give out for free in exchange of a little convenience.