Step 1. Go to support.microsoft.com.
Step 2. Type "convert openoffice".
Step 3. Read the result: " We Currently Have No Documents That Match Your Search"
Many years ago I had a girlfriend who found the clickety-clack
vibration of a train, well, exciting if you know what I mean. One of
her fantasies was to spend a night in a sleeper car. Unfortunately the
opportunity never came up, at least not with me (:
The trust that can be placed in a computer-verified proof is no more and no less than
the trust in the program that
does the verification. That computer program must be correct; if it is,
then the verified proof is a close to being absolutely correct and
rigorous as is humanly and theoretically possible. So that program is the one thing that must be bug-free. Bugs
or errors in the proof itself are impossible
by definition if the
verifier is bug-free. (If a bad proof can slip
by the verifier then that is a bug in the
verifier.)
Some proof-verification languages require verification programs that
are rather large. For example, it has been
estimated that a minimal verifier for HOL (a relatively light
verification language) would require about 3000 lines of Python code for a
verifier. Two projects that have significantly
more modest requirements for their
checking code are Ghilbert
(currently in its early stages) and
Metamath (more mature but with the drawback that new definitions can only be introduced as
new axioms and therefore must
be manually checked for soundness).
A proof verifier
has been written for the latter with only 300 lines of Python code.
Also neat is the Reuleaux Triangle
that is not round but even so has a constant width as it rotates. If it is
used as a roller between two planks, it will roll smoothly and the
distance between the planks will remain constant. This java applet
demonstrates it.
If think Microsoft really promotes good standards, ask Andrew Tridgell (Samba
team leader) who's practically dedicated his life to reverse-engineering
Microsoft's SMB protocol. In this
interview
he says:
"The protocol is so incredibly convoluted and bloated and badly designed
-- there are ten ways of doing everything. You end up with these
massive exchanges going on the wire between Windows 95 and NT, just
because they are trying to work out exactly which sets of bugs the other
guy has so they can figure out how to actually stat a file or find its
size or date or something. And we've found from talking to people who
work at Microsoft how much of a headache it is to maintain the damned
thing and keep it secure."
I love Kedit and it is one of the few things (come to think of it, maybe
the only thing - well, that and TurboTax once a year) that keeps me from
switching full-time to Linux. Every so often I try it with the latest
Wine but no luck yet (it opens, but the screen becomes corrupted after a
while - my gut feel is that it is something simple to fix, and if I ever
find time to work on Wine this would be the one thing I would attack
first - I've even thought of putting up a bounty for getting Wine to
work with Kedit).
I too have requested a Linux port from Mansfield with a cold
response. That company is so infuriating; they own this beautiful thing
and do nothing with it, nor allow anyone else to.
I looked at the THE (The Hessling Editor) project a while back,
but at least then it emulated the DOS version of Kedit. I wish that
project well but the DOS version just isn't the same. Has that
changed? Above I'm talking about KeditW32.
The one valid concern I can see is this, and it's a doozy if you are a record exec...
You go to the company that produced the CD, and they have to
replace it, because they are technically breaking the implicit agreement
that was made when you bought the CD, that you have the right to
personal use of the recorded material. Imagine the record companies
shelling out billions to replace CDs because of natural deterioration.
Have you seen those TV commercials offering ginzu knives or whatever
with a lifetime warranty? The knives are actually cheaply made.
When one goes bad you can send it in and have it replaced FREE, plus
a "modest shipping and handling charge" of $6.95 for a knife
that might cost $0.50 to make. Their "lifetime warranty" has just turned into perpetual guaranteed
income for them. The only requirement in a
scheme like this is that the item be cheap to
produce. You can do the math for CDs.
I recently bought a Compaq Presario 2500 laptop for $850, added 256M of memory to reach
512M (+$60). It has a Celeron 2.4GHz CPU. I am now regretting getting
the Celeron; for another $100 I could have gotten an AMD. This machine
is noticeably more sluggish than the 1GHz Pentium Vaio w/ 256MB it
replaces. rsync'ing from scratch to a large remote directory (10000+
small files in 500MB) under Cygwin takes longer - about 25 vs. 15
minutes - this is a CPU time limitation; the network is local and 100Mb
and the disk is not continuously active.
Re:Readers might also enjoy
on
Everything and More
·
· Score: 4, Interesting
Readers (or rather web surfers) might also enjoy the Metamath web site. Try the "Metamath Proof
Explorer" which "constructs mathematics from scratch" quite literally,
up through Cantor's infinity and beyond.
I trust that timing the announcement of this proposal on the day after
the
Madrid terrorist bombings is just an unfortunate coincidence. Not
that it would affect the public's sentiment one way or the other, right?
And we can be confident that Congress will study it rationally and
objectively, as demonstrated by their carefully considered passage of
the Patriot Act, passed almost unanimously before any single human could
even read all 800 pages of it, much less grasp its scope.
Re:Doing A Hundred Jobs - All Badly
on
USB Swiss Army Knife
·
· Score: 3, Informative
But lets not bag on MS with general crap, this affects Linux too.
Linux has a tremendous advantage that MS doesn't have. Because they
are GPL'd, most major apps can be (and are) included in standard distros
and verified to work in advance by the distro integrators (RedHat, SuSE,
etc.) prior to major releases. It is all released as one big integrated
package where everything works together, and for the most part
everything you'll need is just there. Only now and then you might need
a rarely-used specialty app that isn't included in the standard distro.
In contrast, getting a usable MS system might involve proprietary
programs from a dozen different companies. Every time you upgrade the
MS system, there is no guarantee that none of your third-party apps will
break, and you are the one who has to make sure that they work together.
You may have to test them all, check out each vendor's release notes,
install updates (sometimes for a price) to ensure compatibility, and so
on.
This is also a problem when you have to reinstall Windows from
scratch (without a system recovery disk frozen to a specific
configuration, that may no longer work when you've swapped some
hardware). Installing Windows itself is often the small part of the
project. It might take a good part of a day to install your third-party
apps in a highly unautomated way that requires you to type in CD keys
and so on. People sometimes compare the time to install Linux to the
time needed to install Windows, then ignore the time needed to install
all the Windows apps.
Unfortunately MS is stuck between a rock and a hard place. They
can't integrate in the functionality of all the third-party apps that
Linux distros have by default without antitrust accusations.
The bill's biggest backers are the Software and Information Industry
Association; Reed Elsevier, which owns the LexisNexis database; and
Westlaw, the biggest publisher of legal databases.
So I guess the (high priced) LexisNexis feels threatened
by the free information provided by Google et. al.
An encyclopedia site not only could own the historical facts contained
in its online entries, but could do so long after the copyright on
authorship of the written entries had expired. Unlike copyright, which
expires 70 years after the death of a work's author, the
Misappropriation Act doesn't designate an expiration date.
Finally we're getting perpetual ownership of information. It's only
a matter of time before it gets put into regular copyrights in order to
harmonize the laws. Disney's wet dream come true.
Commercial database companies say they invest millions of dollars in
collecting, editing and organizing information for their customers, but
don't have adequate protection to prevent someone from stealing the
information to compete with them.
To me this is the worst possible justification for a new law.
No one made them invest millions of dollars.
Bill Gates is a multi-billionaire,
Linus Torvalds is a multi-thousandaire
Although it is NOMB, I would guess Linus is at least a multi-millionaire
(meaning >= 2 million). He was given a big chunk of RedHat stock or
options at the IPO, and I recall reading that he shrugged his shoulders and said something
like, "I guess I'm rich now." I think that was before RHAT's insane
peak. Don't know if he cashed any out near the peak. I myself
qualified for those 400 shares at the IPO price given to open source
developers and did well. I'm sure Linus had
much much more than that.
While VR pain relief may work to some degree initially, once the
novelty wears off, or on an off-day when you just can't get interested
in its "game", you'll probably find yourself screaming with pain.
Should I ever find myself in such an unfortunate situation, may God have
mercy on me and set me up with an MD who will prescribe adequate opioid
pain relievers. Currently that is the only thing that works, period.
Too often these days MDs are paranoid about prescribing opioid pain
killers, what with the DEA breathing down their necks. See The DEA's
Disastrous War Against Pain-Treating Drugs for
example. It is customary to
encourage the patient to grin and bear it or to seek pain relief through
alternative therapies like meditation etc.
I myself have had minor surgery were they'll give you plenty of local
anesthetic during the actual procedure; then they send you home with
instructions to take tylenol. When the anesthetic wears off, the pain
kicks in. It is only by whining and complaining that they'll prescribe
an opioid painkiller, and unless you go to the ER (and sometimes even if you
do) you'll be in pain for hours more until all the paperwork and
procedures are done to get the prescription filled.
Chronic pain patients are in a real bind these days.
They cruelty towards them by denying them long-term opioid pain
relief is unspeakable.
Good luck getting contractors to write code for the gov't then...
These days I think they'll have a lot of luck
getting contractors to write code for the gov't.
Besides, why is it any different from any other
work for hire?
And as an employee of said contractor, who
wouldn't have any copyright interest in whatever I
produce anyway, I think I might be more motivated to
produce better work if I knew it would ultimately
be subject to public scrutiny and benefit the
public good. Compare that to dedicating your
life to writing code that will be secreted away
in some closed-source product with no
acknowledgement whatsoever to you other than
a paycheck that lets you survive. The thought
of such a dismal and pointless existence is
kind of depressing.
According to copyright law, you're paying for the right to make a
copy, not for the right to listen.
I think it's more complicated than that. You cannot take your copy,
that you purchased and legally own, and play it for a large audience
though there is no further copying involved.
I purchased well over 50 CDs in my lifetime. I get back $13? From
my quick calculations I feel that I should be getting back about $300
instead.
You're doing the math wrong. Here are the equations you should use;
it's actually quite simple:
If you are a customer and (potentially) screw a record company by
infringing a copyright, you owe $150,000 times the number of
incidents.
If you are a record company and (actually) screw a customer by
illegally overcharging, you owe the customer $13 times the number
of incidents, then divided by the total by the number of incidents.
I tried moving a customer's aging win NT sql server over to MySQL on red
hat to save them some money, but as it turned out the people who wrote
the billing software that used their SQL server refused to allow them to
use anything but MS-SQL 2000 (claiming it MySQL didn't support the
functions they used).
At the risk of starting a flame war, I think you should have
suggested PostgreSQL, especially for billing software. It has a cleaner
implementation of standard SQL and transactions that I think would
appeal better to billing software types. It might have supported the
functions they used, although you don't say what they were.
I converted an Oracle project to Postgres relatively painlessly.
In particular, the language used for stored procedures is very similar
and can almost by translated algorithmicly. (Actually I wrote a
quick and dirty translator for some of the more routine things.)
While Postgres might not handle high-volume mainframe stuff as efficiently
as Oracle - the data is unclear on this - certainly it's more
than adequate as a replacement for anything running on an
"aging win NT sql server".
But there is NEVER a reason to shut down a Windows XP computer (if
you're not installing anything or changing settings). My computer has
been on for 92 days and is still as stable and fast as it was on day 1
(super fast and 100% stable).
Since you don't install the security updates (which require a reboot)
I certainly hope your computer isn't connected to the Internet.
Otherwise it's probably been turned into a zombied relay for spammers
who are all too happy with your 92 day uptime.
If this is true then Habeas is really screwing themselves. The only
thing Habeas has got going for it is its reputation, and if it does not
aggressively address complaints
about its customers who
violate their policy, it's reputation will become ruined. Let's see -
there's your report, there are the spammers they are currently
pursuing, and there is the worrisome "Easy to defeat...." earlier post
by DrPepper about Joe-Jobs (now that's clever; I wonder if some spammer will
patent it). More and more it seems the Habeas headers are becoming
something to blacklist rather than whitelist.
Has anyone ever found whitelisting Habeas useful? It sounds like
it's supposed to be a way to receive just barely tolerable opt-out-able
advertisements. The benefit to me
seems marginal, and by whitelisting them I'm doing them a favor by agreeing
to receive the "polite" advertisements of their customers. It won't
take much to push me over the edge and blacklist them. I
have to take your post at face value and
have no way to verify what you say, but it is
still enough to push me closer to that edge.
Someone from Habeas should respond to the parent post. Otherwise
they're going into my blacklist. So far I have received nothing useful
through their service that I can recall, and I don't need more
advertisements, not even "non-spam" ones,
especially if they ignore complaints like yours.
My company bought an eMachines 333cs from 1999. It was cheap, $400. It came with W98,
and when we put NT on it locked up every few days for no good reason.
Later I heard this was a frequent problem, and
this is probably one of the reasons the machine
was considered crap. Anyway around 2001(? or
whenever 7.1 came out) I put RH7.1 on it (stripped down to bare bones
for security), rebuilt the kernel to 2.4 so I could do iptables and
iproute2, and configured it to be an ARP proxy transparent
firewall/router for our office, which it still is and will continue to
be for the foreseeable future.
I remember when Code Red came out I ssh'ed in
from home (actually to an internal Linux machine,
then to the firewall - the firewall has all ports
blocked to the outside) and blocked it. Remote
admin is good.
The power supply fan died twice (once burned out, once frozen) and
the CPU fan froze once. I now have a couple of spares of these fans for
a few years to come. I have a streamer on the power supply fan to
show me it's working. Except for turning it off to replace the fans, the
uptime has been continuous, about 1.5 years until the first fan problem.
Lasting 50-100 times longer still won't break even, at the Radio Shack
price. Each LED consumes 3.6v * 20ma = 0.07W. If we are generous and
assume that they are 5 times as efficient as incandescent, like a
fluorescent is - actually, I don't think they are anywhere near that,
certainly not if measured in terms of total lumens - then we would need
20W of them to equal a 100W incandescent, i.e. we would need 285 of
them. At the $4.99 Radio Shack price, that would be $1422.15 to get the
same light as a $0.99 incandescent bulb.
I'm sure if you shop around, though, you'll be able to do better
than the Radio Shack price, especially if you're going to buy
285 of them for each bulb in your house you're going to replace.
The real question (assuming money is no object) is whether the Radio
Shack LED is really white. All of the so-called "white" LEDs I've
purchased, from the Kensington USB light for my laptop to the little
flashlight on my keychain, have an annoying purplish/blue tinge that
distorts colors and (for reading) gets tiring after a while. I've
switched my booklight for reading in bed back to an incandescent one.
From the pictures of the LED Apartment, apparently they have the same
problem with most of their LEDs, too.
Slashdot Mirror
Step 1. Go to support.microsoft.com.
Step 2. Type "convert openoffice".
Step 3. Read the result: " We Currently Have No Documents That Match Your Search"
Many years ago I had a girlfriend who found the clickety-clack vibration of a train, well, exciting if you know what I mean. One of her fantasies was to spend a night in a sleeper car. Unfortunately the opportunity never came up, at least not with me (:
Some proof-verification languages require verification programs that are rather large. For example, it has been estimated that a minimal verifier for HOL (a relatively light verification language) would require about 3000 lines of Python code for a verifier. Two projects that have significantly more modest requirements for their checking code are Ghilbert (currently in its early stages) and Metamath (more mature but with the drawback that new definitions can only be introduced as new axioms and therefore must be manually checked for soundness). A proof verifier has been written for the latter with only 300 lines of Python code.
Also neat is the Reuleaux Triangle that is not round but even so has a constant width as it rotates. If it is used as a roller between two planks, it will roll smoothly and the distance between the planks will remain constant. This java applet demonstrates it.
This, my friend, is a Microsoft "standard".
I too have requested a Linux port from Mansfield with a cold response. That company is so infuriating; they own this beautiful thing and do nothing with it, nor allow anyone else to.
I looked at the THE (The Hessling Editor) project a while back, but at least then it emulated the DOS version of Kedit. I wish that project well but the DOS version just isn't the same. Has that changed? Above I'm talking about KeditW32.
You go to the company that produced the CD, and they have to replace it, because they are technically breaking the implicit agreement that was made when you bought the CD, that you have the right to personal use of the recorded material. Imagine the record companies shelling out billions to replace CDs because of natural deterioration.
Have you seen those TV commercials offering ginzu knives or whatever with a lifetime warranty? The knives are actually cheaply made. When one goes bad you can send it in and have it replaced FREE, plus a "modest shipping and handling charge" of $6.95 for a knife that might cost $0.50 to make. Their "lifetime warranty" has just turned into perpetual guaranteed income for them. The only requirement in a scheme like this is that the item be cheap to produce. You can do the math for CDs.
I recently bought a Compaq Presario 2500 laptop for $850, added 256M of memory to reach 512M (+$60). It has a Celeron 2.4GHz CPU. I am now regretting getting the Celeron; for another $100 I could have gotten an AMD. This machine is noticeably more sluggish than the 1GHz Pentium Vaio w/ 256MB it replaces. rsync'ing from scratch to a large remote directory (10000+ small files in 500MB) under Cygwin takes longer - about 25 vs. 15 minutes - this is a CPU time limitation; the network is local and 100Mb and the disk is not continuously active.
Readers (or rather web surfers) might also enjoy the Metamath web site. Try the "Metamath Proof Explorer" which "constructs mathematics from scratch" quite literally, up through Cantor's infinity and beyond.
I trust that timing the announcement of this proposal on the day after the Madrid terrorist bombings is just an unfortunate coincidence. Not that it would affect the public's sentiment one way or the other, right? And we can be confident that Congress will study it rationally and objectively, as demonstrated by their carefully considered passage of the Patriot Act, passed almost unanimously before any single human could even read all 800 pages of it, much less grasp its scope.
Wenger does offer one with a laser pointer built in. For a comparison of the two companies, see http://outside.away.com/magazine/200007/200007disp 3.html.
Linux has a tremendous advantage that MS doesn't have. Because they are GPL'd, most major apps can be (and are) included in standard distros and verified to work in advance by the distro integrators (RedHat, SuSE, etc.) prior to major releases. It is all released as one big integrated package where everything works together, and for the most part everything you'll need is just there. Only now and then you might need a rarely-used specialty app that isn't included in the standard distro.
In contrast, getting a usable MS system might involve proprietary programs from a dozen different companies. Every time you upgrade the MS system, there is no guarantee that none of your third-party apps will break, and you are the one who has to make sure that they work together. You may have to test them all, check out each vendor's release notes, install updates (sometimes for a price) to ensure compatibility, and so on.
This is also a problem when you have to reinstall Windows from scratch (without a system recovery disk frozen to a specific configuration, that may no longer work when you've swapped some hardware). Installing Windows itself is often the small part of the project. It might take a good part of a day to install your third-party apps in a highly unautomated way that requires you to type in CD keys and so on. People sometimes compare the time to install Linux to the time needed to install Windows, then ignore the time needed to install all the Windows apps.
Unfortunately MS is stuck between a rock and a hard place. They can't integrate in the functionality of all the third-party apps that Linux distros have by default without antitrust accusations.
So I guess the (high priced) LexisNexis feels threatened by the free information provided by Google et. al.
An encyclopedia site not only could own the historical facts contained in its online entries, but could do so long after the copyright on authorship of the written entries had expired. Unlike copyright, which expires 70 years after the death of a work's author, the Misappropriation Act doesn't designate an expiration date.
Finally we're getting perpetual ownership of information. It's only a matter of time before it gets put into regular copyrights in order to harmonize the laws. Disney's wet dream come true.
Commercial database companies say they invest millions of dollars in collecting, editing and organizing information for their customers, but don't have adequate protection to prevent someone from stealing the information to compete with them.
To me this is the worst possible justification for a new law. No one made them invest millions of dollars.
Although it is NOMB, I would guess Linus is at least a multi-millionaire (meaning >= 2 million). He was given a big chunk of RedHat stock or options at the IPO, and I recall reading that he shrugged his shoulders and said something like, "I guess I'm rich now." I think that was before RHAT's insane peak. Don't know if he cashed any out near the peak. I myself qualified for those 400 shares at the IPO price given to open source developers and did well. I'm sure Linus had much much more than that.
Anyway that's just speculation on my part.
While VR pain relief may work to some degree initially, once the novelty wears off, or on an off-day when you just can't get interested in its "game", you'll probably find yourself screaming with pain.
Should I ever find myself in such an unfortunate situation, may God have mercy on me and set me up with an MD who will prescribe adequate opioid pain relievers. Currently that is the only thing that works, period.
Too often these days MDs are paranoid about prescribing opioid pain killers, what with the DEA breathing down their necks. See The DEA's Disastrous War Against Pain-Treating Drugs for example. It is customary to encourage the patient to grin and bear it or to seek pain relief through alternative therapies like meditation etc.
I myself have had minor surgery were they'll give you plenty of local anesthetic during the actual procedure; then they send you home with instructions to take tylenol. When the anesthetic wears off, the pain kicks in. It is only by whining and complaining that they'll prescribe an opioid painkiller, and unless you go to the ER (and sometimes even if you do) you'll be in pain for hours more until all the paperwork and procedures are done to get the prescription filled.
Chronic pain patients are in a real bind these days. They cruelty towards them by denying them long-term opioid pain relief is unspeakable.
These days I think they'll have a lot of luck getting contractors to write code for the gov't. Besides, why is it any different from any other work for hire?
And as an employee of said contractor, who wouldn't have any copyright interest in whatever I produce anyway, I think I might be more motivated to produce better work if I knew it would ultimately be subject to public scrutiny and benefit the public good. Compare that to dedicating your life to writing code that will be secreted away in some closed-source product with no acknowledgement whatsoever to you other than a paycheck that lets you survive. The thought of such a dismal and pointless existence is kind of depressing.
I think it's more complicated than that. You cannot take your copy, that you purchased and legally own, and play it for a large audience though there is no further copying involved.
You're doing the math wrong. Here are the equations you should use; it's actually quite simple:
If you are a customer and (potentially) screw a record company by infringing a copyright, you owe $150,000 times the number of incidents.
If you are a record company and (actually) screw a customer by illegally overcharging, you owe the customer $13 times the number of incidents, then divided by the total by the number of incidents.
You left out the denominator.
View -> Message Body As -> Plain Text
At the risk of starting a flame war, I think you should have suggested PostgreSQL, especially for billing software. It has a cleaner implementation of standard SQL and transactions that I think would appeal better to billing software types. It might have supported the functions they used, although you don't say what they were.
I converted an Oracle project to Postgres relatively painlessly. In particular, the language used for stored procedures is very similar and can almost by translated algorithmicly. (Actually I wrote a quick and dirty translator for some of the more routine things.) While Postgres might not handle high-volume mainframe stuff as efficiently as Oracle - the data is unclear on this - certainly it's more than adequate as a replacement for anything running on an "aging win NT sql server".
Uhm, I didn't know w3m had a logo. In any case it wouldn't render on the browser itself, so how would people using w3m see the logo?
Oh, now I get it. I bet it's ASCII art! Can you point me to one of your sites?
Ah, by chance you didn't mean the w3c logo, did you?
Since you don't install the security updates (which require a reboot) I certainly hope your computer isn't connected to the Internet. Otherwise it's probably been turned into a zombied relay for spammers who are all too happy with your 92 day uptime.
Has anyone ever found whitelisting Habeas useful? It sounds like it's supposed to be a way to receive just barely tolerable opt-out-able advertisements. The benefit to me seems marginal, and by whitelisting them I'm doing them a favor by agreeing to receive the "polite" advertisements of their customers. It won't take much to push me over the edge and blacklist them. I have to take your post at face value and have no way to verify what you say, but it is still enough to push me closer to that edge.
Someone from Habeas should respond to the parent post. Otherwise they're going into my blacklist. So far I have received nothing useful through their service that I can recall, and I don't need more advertisements, not even "non-spam" ones, especially if they ignore complaints like yours.
My company bought an eMachines 333cs from 1999. It was cheap, $400. It came with W98, and when we put NT on it locked up every few days for no good reason. Later I heard this was a frequent problem, and this is probably one of the reasons the machine was considered crap. Anyway around 2001(? or whenever 7.1 came out) I put RH7.1 on it (stripped down to bare bones for security), rebuilt the kernel to 2.4 so I could do iptables and iproute2, and configured it to be an ARP proxy transparent firewall/router for our office, which it still is and will continue to be for the foreseeable future.
I remember when Code Red came out I ssh'ed in from home (actually to an internal Linux machine, then to the firewall - the firewall has all ports blocked to the outside) and blocked it. Remote admin is good.
The power supply fan died twice (once burned out, once frozen) and the CPU fan froze once. I now have a couple of spares of these fans for a few years to come. I have a streamer on the power supply fan to show me it's working. Except for turning it off to replace the fans, the uptime has been continuous, about 1.5 years until the first fan problem.
I'm sure if you shop around, though, you'll be able to do better than the Radio Shack price, especially if you're going to buy 285 of them for each bulb in your house you're going to replace.
The real question (assuming money is no object) is whether the Radio Shack LED is really white. All of the so-called "white" LEDs I've purchased, from the Kensington USB light for my laptop to the little flashlight on my keychain, have an annoying purplish/blue tinge that distorts colors and (for reading) gets tiring after a while. I've switched my booklight for reading in bed back to an incandescent one. From the pictures of the LED Apartment, apparently they have the same problem with most of their LEDs, too.