> The article is talking about stuff that's available today. They aren't saying "SSDs will never be suitable", they're saying they aren't suitable today.
They are suitable today. You just don't raid them using 50-100K RAID controllers.
Anyway, the "Enterprise Storage" bunch will probably stick both SSDs and TB SATA drives in their systems for the speed and capacity (and charge $$$$$$). I think some are doing it already.
Or you could stick a few SSDs in a decent x86 server with 10 Gbps NICs, and now you can have the same amount of IOPS as you would after spending > 100K on RAID controllers and drives.
Not so sure about hotswapping SSDs though - so far I don't see much info on that;).
I think almost everything alive can feel pain. Whether they have the capacity to tell us is a different matter.
If you are experiencing "locked in syndrome", you're not more capable than an amoeba. If I poke you you might feel pain, but what can you do about it? Nothing. At least the amoeba can move away.
As for eating animals, there's plenty of scientific evidence that consuming oceanic fish (nonmercury laden) is good for most humans (all that omega 3 stuff). And consuming red meat is generally not. There are always exceptions of course. So while humans can survive on a pure vegetarian diet, it's hard for them to thrive on it. They are less likely to do as well as those who eat fish.
Unfortunately most vegetarian sources of Omega 3 oils don't provide the long chained ones (DHA, EPA), they provide ALA and:
"More specifically, most studies in humans have shown that whereas a certain, though restricted, conversion of high doses of ALA to EPA occurs, conversion to DHA is severely restricted."
> Finally, if at all possible, make sure your DHCP server sends ACK using unicast where possible. AFAIK, every major OS should be able to handle this
Windows Vista by default (stupidly/evilly pick one) sets the broadcast flag for DHCP, so it requires broadcast responses. While it can handle unicast replies if configured accordingly, by default it doesn't ask for unicast replies and thus should not be getting unicast replies.
That said, "default Vista" is OK with "broadcast replies" which are are unicast layer 2 (dest= mac address of vista machine) but broadcast layer 3 (dest=255.255.255.255). This may or may not be RFC compliant (in my reading the RFC is not very specific on this), but it works.
How I know this? Because in my previous workplace (which supplied "expensive hotel/airport internet") I wrote a dhcp server which was somewhat RFC compliant (and did work with Vista unlike some other dhcp servers;) ), but I had to deal with a scenario where some network devices between the client and server were not forwarding layer 2 broadcast frames (they were supposed to) - so the dhcp replies never reached the vista clients. Fixing the devices in time was not possible, so I worked around it by doing the above.
If 802.11 broadcasts are that expensive (I can see why it would be different from unicast but are there any decent articles on that? ), then it looks like that feature would be useful in this scenario too - since you then send a 802.11 unicast but have the dest IP = 255.255.255.255.:)
Not sure if some patent troll has patented that already - to me it's just an obvious solution.
I don't see a good technical reason why Vista (or other DHCP client) should use broadcasts by default though.
And I wonder though how many laptop users are actually using Vista and sticking with it, instead of moving to Windows 7?
> And then again in five minutes, and five minutes after that
Doesn't seem to do that on my Windows boxes though and I run XP and Windows 7. The only annoyance I have is with Windows 7 which still installs the updates when I shutdown, even though they are supposed to only install when I say so.
Despite that annoyance you're talking about seems lots of Windows users still don't update their machines often enough:).
There are plenty of meetings where the participation is not inconsequential[1], the trouble is only a few minutes of it is important (where you take part actively or have to really focus), the other X hours don't require your full concentration.
In those meetings the participants only need to concentrate when it's their section of the agenda, but they still need to be in the meeting to keep up to date with what the participants are doing and discussing, but that does not need much concentration (unless of course basic reading and comprehension is too hard, in which case you have a different problem [2] ).
> The concept of a meeting is not broken or inefficient; the application of meetings to problems/people that aren't best solved by a meeting is.
Sure, but most people who get the power to hold meetings still don't know when to hold meetings and who should be in them, and that sure looks like it isn't going to change.
On the other hand, the upcoming generation of people that are using IM a lot (unlike their parents and grandparents) might be more acceptable to the sort of meetings I talk about. And if such meetings do become commonplace, the inefficiencies go down, even if people hold meetings unnecessarily.
> Nothing is more frustrating than trying to deal with someone who you KNOW is distracted. It's rude, and its inefficient.
With my suggestion, if a person is regularly a bottleneck, the bosses can see it in the archived chat logs of the meetings - the timestamps are plain to see.
There are tasks which require full concentration - most people don't code/drive/conduct surgery/etc well when distracted. But in my experience participating in meetings hardly ever requires full sustained concentration by most of the participants. So participating in just one meeting at a time is a waste of their time. A boss wanting to squeeze out more productivity would thus encourage what I'm talking about.
From a personal point of view I'm actually ambivalent about my suggestion - bosses squeezing every last drop of productivity out of employees isn't always such a great thing, but the topic is inefficiency after all, and there's plenty of inefficiencies about and this is one of them.
I personally do not think that Productivity and Efficiency should be like Gods to us. To me the end goal of productivity and efficiency should be a better life for most people. If higher productivity and efficiency only results in a better life for just the "upper classes", then what we have achieved isn't much better than slavery in the old days.
[1] There have been some meetings which officially I probably shouldn't have been in - turns out it's not really my scope, but I still managed to contribute some rather useful points. Similarly there are lots of people who still make very useful posts to Slashdot on topics that aren't in their official fields of study or work.
[2] Analogy: If you have weak CPUs multitasking tasks can be counterproductive. But if you have powerful CPUs, while there are scheduling overheads, overall you can squeeze more out of your powerful CPUs if you get them to multitask, and even then they could still be idle most of the time.
> The reason capitalism and a private sector is so good is that it promotes economic diversity,
But capitalism isn't about promoting economic diversity. It sure seems to tend towards monopolies given half the chance. That's why there are antitrust laws in some countries.
> government regulations and socialism promote a monoculture approach
Depends on the Government regulations. As mentioned there's that anti-trust regulation thing.
The private sector isn't that efficient either. Sure there are examples of efficient companies. But I bet there are also efficient government departments.
Speaking of embracing technology and doing things in new ways, how many companies in the private sector have bosses who encourage meetings (especially internal ones) to be done using instant messaging/IRC?
This increases productivity since employees can be in more than one meeting at the same time, and they can still do other stuff. They could even go to the toilet or answer phones without interrupting the meeting - they can just scroll up when they get back, rather than everyone having the wait for you to get back up to speed.
Whereas physical meetings tend to be very inefficient. A typical meeting could occupy 2 hours of real time from each participant but of which say only 5 minutes are useful.
The rest is "idle time" - wasted. Multiply that by the number of participants and you are looking at a lot of wasted time.
Maybe the next generation would be more accepting of this. But you have to be able to read well and fast enough.
I'm not saying physical face to face meetings should go away completely - there will still be good reasons for them, but for so many meetings (especially more technical ones) they are unnecessary (given suitable IM software).
But I don't want it to install updates just because I shutdown...
I want it to install updates when I choose to install them.
Anyway I don't see a big difference between Windows and say Ubuntu with this patching stuff.
1) Both require regular updates 2) In most normal cases both require restarting of software that's updated whether it's the O/S or apps or services. Believe me if you update Open Office you need to restart it for the update to take effect.
The Linux users who are smugly not restarting stuff are just running the out of date versions.
A restart is a simple way to be certain that the "Aunt May" and "Joe Sixpack" bunch are no longer running the unpatched stuff.
It seems some ignorant Linux fans (especially those not using ksplice) should be rebooting their systems more often than they are doing...
If you aren't rebooting after a linux kernel update, you are likely to be still running the old kernel with bugs.
Yes there are ways to update a kernel "in-place"[1] but I doubt your distro does that yet by _default_.
Lots of things require a restart when they are updated for the updates to take effect.
If you update apache or your GUI, you need to restart them.
Yes you can choose to not restart the stuff, but you are then NOT running the updated stuff.
Windows XP/7 doesn't force you to restart either. A restart is a simple way to ensure you are not running the out of date stuff.
[1] http://www.linux-mag.com/cache/7403/1.html
I'm not sure if that tech works with stuff like vmware. I always seem to need to do some stuff after kernel updates for vmware to work again (google for details)...
But that doesn't really sound like a Windows-specific problem to me.
That problem could happen with Linux too.
There's also lots of half-baked advice out there that tell users to delete certain stuff in windows\... while that frees up space, things like "windows update" can stop working.
And as we have seen, many users get infested with malware, and so updates cause problems (like BSODs).
In most cases it's better to buy pre-crimped certified cables and test them, than to crimp them and test them.
How much money are you going to save? If you're charging $$$ for the project, crimping your own cables is a waste of time and resources. Unless of course your job is making cables:).
b (bits) and s (second) don't work in all languages either.
We spend lots of time complaining about people who confuse bps vs Bps, and milli with Mega, and there are the kibibyte vs kilobyte wars. And now b/s vs bps?
But you couldn't control exactly what was written with the Apple II drives.
IIRC the spec was you could not expect to _reliably_ write a bit stream that has more than two zeroes in a row. In practice you could usually get away with three zeroes in a row. You had to do stuff like write sync streams e.g. 11111111001111111100111111110011111111001111111100. Which are basically 5 x 10 bits of 1111111100.
You could control the drive head too. You could even move the head in sub-track increments - e.g. half track. However, tracks have to be at least a full track apart to be reliably read/written.
All these is why I find the submitter's request rather strange. When you get down to the low level details, there is no such thing as exact.
Even if you are writing one bit on the entire platter with a huge magnet:).
So I'm curious on what he's really trying to achieve.
Just curious: are you satisfied with what you were getting for your money?
Would you prefer to pay less and have BBC make these cuts?
FWIW, the BBC is still one of the most respected media organizations in the world.
Unfortunately it's not very capitalistic or "free market"-ish to force everyone with a TV (or equivalent) to pay for the BBC. But if the alternative is more Rupert Murdoch and Friends...
But what if all the attacker needs is for you to click on the link in the email?
e.g. overworked admin gets an email that appears to be from "That idiot ISP who keeps sending me announcements via http/pdf/msword links".
Admin clicks on link, gets pwned without knowing.
If the admin has html email enabled (just to be able to read instructions from bosses )the whole message could be a link too, so you could still accidentally click on the payload even if you know it is suspicious (e.g. misclick while trying to confirm the headers...).
The attackers may not need to resort to this yet since there are enough gullible people around. But looking at the number of drive-by exploits (whether for Windows, Linux or OSX) it doesn't give you much confidence that you won't ever get pwned.
Sorry about the jab back - I shouldn't, but just couldn't resist.
> Is there any actual study, however, which tested how "accessible" placebo is to atheists vs. believers? Much of the effectiveness, from what I've read, relies on the person believing that the (fake) treatment works.
My reasoning is that some believers can just pray to their God wherever they are. And even if their God doesn't actually help (because of nonexistence or other) they could still experience the placebo effect.
In contrast, atheists could try praying to the FSM, I currently doubt that's going to work as well.
HOWEVER I may be wrong since that has not been tested (to my knowledge). Perhaps the relevant parts of the brain could still work to produce the placebo effect when you pray to X even if you don't believe that X exists. That would be an interesting result. However I'm not sure if any reputable scientists are going to conduct such a study.
While dogs might actually believe in supernatural stuff, I doubt religion is required for the placebo effect, so one could not conclude that dogs are getting religion because placebos work on them.
To figure out whether dogs are getting religion one would have to conduct different experiments.
From the article at least one neighbour didn't mind.
Heck he even helped them:
Soon after the city complained about the yard, the Has placed wood chips on top of the dirt, with help from neighbor Dennis Cleek.
"It's their yard, it's not overgrown with weeds, it's not an eyesore," said Cleek, whose own yard boasts fruit trees. "We should be able to have our yards look the way we want them to."
And from the pic, it looks ok to me. As for wood-chips being a fire hazard, it's no big deal, before they start burning in a dangerous way due to some external cause, those wooden houses will probably be on fire first...
Slashdot Mirror
> The article is talking about stuff that's available today. They aren't saying "SSDs will never be suitable", they're saying they aren't suitable today.
;).
They are suitable today. You just don't raid them using 50-100K RAID controllers.
Anyway, the "Enterprise Storage" bunch will probably stick both SSDs and TB SATA drives in their systems for the speed and capacity (and charge $$$$$$). I think some are doing it already.
Or you could stick a few SSDs in a decent x86 server with 10 Gbps NICs, and now you can have the same amount of IOPS as you would after spending > 100K on RAID controllers and drives.
Not so sure about hotswapping SSDs though - so far I don't see much info on that
I think almost everything alive can feel pain. Whether they have the capacity to tell us is a different matter.
If you are experiencing "locked in syndrome", you're not more capable than an amoeba. If I poke you you might feel pain, but what can you do about it? Nothing. At least the amoeba can move away.
As for eating animals, there's plenty of scientific evidence that consuming oceanic fish (nonmercury laden) is good for most humans (all that omega 3 stuff). And consuming red meat is generally not. There are always exceptions of course. So while humans can survive on a pure vegetarian diet, it's hard for them to thrive on it. They are less likely to do as well as those who eat fish.
Unfortunately most vegetarian sources of Omega 3 oils don't provide the long chained ones (DHA, EPA), they provide ALA and:
http://www.ncbi.nlm.nih.gov/pubmed/9637947
"More specifically, most studies in humans have shown that whereas a certain, though restricted, conversion of high doses of ALA to EPA occurs, conversion to DHA is severely restricted."
Seems more like self-waterboarding...
:).
You'd invent almost anything just to not drown
> Finally, if at all possible, make sure your DHCP server sends ACK using unicast where possible. AFAIK, every major OS should be able to handle this
Windows Vista by default (stupidly/evilly pick one) sets the broadcast flag for DHCP, so it requires broadcast responses. While it can handle unicast replies if configured accordingly, by default it doesn't ask for unicast replies and thus should not be getting unicast replies.
See: http://support.microsoft.com/kb/928233
Windows 7 probably handles it better: http://blogs.technet.com/teamdhcp/archive/2009/02/12/dhcp-broadcast-flag-handling-in-windows-7.aspx
That said, "default Vista" is OK with "broadcast replies" which are are unicast layer 2 (dest= mac address of vista machine) but broadcast layer 3 (dest=255.255.255.255). This may or may not be RFC compliant (in my reading the RFC is not very specific on this), but it works.
How I know this? Because in my previous workplace (which supplied "expensive hotel/airport internet") I wrote a dhcp server which was somewhat RFC compliant (and did work with Vista unlike some other dhcp servers ;) ), but I had to deal with a scenario where some network devices between the client and server were not forwarding layer 2 broadcast frames (they were supposed to) - so the dhcp replies never reached the vista clients. Fixing the devices in time was not possible, so I worked around it by doing the above.
If 802.11 broadcasts are that expensive (I can see why it would be different from unicast but are there any decent articles on that? ), then it looks like that feature would be useful in this scenario too - since you then send a 802.11 unicast but have the dest IP = 255.255.255.255. :)
Not sure if some patent troll has patented that already - to me it's just an obvious solution.
I don't see a good technical reason why Vista (or other DHCP client) should use broadcasts by default though.
And I wonder though how many laptop users are actually using Vista and sticking with it, instead of moving to Windows 7?
> And then again in five minutes, and five minutes after that
:).
Doesn't seem to do that on my Windows boxes though and I run XP and Windows 7. The only annoyance I have is with Windows 7 which still installs the updates when I shutdown, even though they are supposed to only install when I say so.
Despite that annoyance you're talking about seems lots of Windows users still don't update their machines often enough
There are plenty of meetings where the participation is not inconsequential[1], the trouble is only a few minutes of it is important (where you take part actively or have to really focus), the other X hours don't require your full concentration.
In those meetings the participants only need to concentrate when it's their section of the agenda, but they still need to be in the meeting to keep up to date with what the participants are doing and discussing, but that does not need much concentration (unless of course basic reading and comprehension is too hard, in which case you have a different problem [2] ).
> The concept of a meeting is not broken or inefficient; the application of meetings to problems/people that aren't best solved by a meeting is.
Sure, but most people who get the power to hold meetings still don't know when to hold meetings and who should be in them, and that sure looks like it isn't going to change.
On the other hand, the upcoming generation of people that are using IM a lot (unlike their parents and grandparents) might be more acceptable to the sort of meetings I talk about. And if such meetings do become commonplace, the inefficiencies go down, even if people hold meetings unnecessarily.
> Nothing is more frustrating than trying to deal with someone who you KNOW is distracted. It's rude, and its inefficient.
With my suggestion, if a person is regularly a bottleneck, the bosses can see it in the archived chat logs of the meetings - the timestamps are plain to see.
There are tasks which require full concentration - most people don't code/drive/conduct surgery/etc well when distracted. But in my experience participating in meetings hardly ever requires full sustained concentration by most of the participants. So participating in just one meeting at a time is a waste of their time. A boss wanting to squeeze out more productivity would thus encourage what I'm talking about.
From a personal point of view I'm actually ambivalent about my suggestion - bosses squeezing every last drop of productivity out of employees isn't always such a great thing, but the topic is inefficiency after all, and there's plenty of inefficiencies about and this is one of them.
I personally do not think that Productivity and Efficiency should be like Gods to us. To me the end goal of productivity and efficiency should be a better life for most people. If higher productivity and efficiency only results in a better life for just the "upper classes", then what we have achieved isn't much better than slavery in the old days.
[1] There have been some meetings which officially I probably shouldn't have been in - turns out it's not really my scope, but I still managed to contribute some rather useful points. Similarly there are lots of people who still make very useful posts to Slashdot on topics that aren't in their official fields of study or work.
[2] Analogy: If you have weak CPUs multitasking tasks can be counterproductive. But if you have powerful CPUs, while there are scheduling overheads, overall you can squeeze more out of your powerful CPUs if you get them to multitask, and even then they could still be idle most of the time.
> The reason capitalism and a private sector is so good is that it promotes economic diversity,
But capitalism isn't about promoting economic diversity. It sure seems to tend towards monopolies given half the chance. That's why there are antitrust laws in some countries.
> government regulations and socialism promote a monoculture approach
Depends on the Government regulations. As mentioned there's that anti-trust regulation thing.
The private sector isn't that efficient either. Sure there are examples of efficient companies. But I bet there are also efficient government departments.
Speaking of embracing technology and doing things in new ways, how many companies in the private sector have bosses who encourage meetings (especially internal ones) to be done using instant messaging/IRC?
This increases productivity since employees can be in more than one meeting at the same time, and they can still do other stuff. They could even go to the toilet or answer phones without interrupting the meeting - they can just scroll up when they get back, rather than everyone having the wait for you to get back up to speed.
Whereas physical meetings tend to be very inefficient. A typical meeting could occupy 2 hours of real time from each participant but of which say only 5 minutes are useful.
The rest is "idle time" - wasted. Multiply that by the number of participants and you are looking at a lot of wasted time.
Maybe the next generation would be more accepting of this. But you have to be able to read well and fast enough.
I'm not saying physical face to face meetings should go away completely - there will still be good reasons for them, but for so many meetings (especially more technical ones) they are unnecessary (given suitable IM software).
But I don't want it to install updates just because I shutdown...
I want it to install updates when I choose to install them.
Anyway I don't see a big difference between Windows and say Ubuntu with this patching stuff.
1) Both require regular updates
2) In most normal cases both require restarting of software that's updated whether it's the O/S or apps or services. Believe me if you update Open Office you need to restart it for the update to take effect.
The Linux users who are smugly not restarting stuff are just running the out of date versions.
A restart is a simple way to be certain that the "Aunt May" and "Joe Sixpack" bunch are no longer running the unpatched stuff.
It seems some ignorant Linux fans (especially those not using ksplice) should be rebooting their systems more often than they are doing...
If you aren't rebooting after a linux kernel update, you are likely to be still running the old kernel with bugs.
Yes there are ways to update a kernel "in-place"[1] but I doubt your distro does that yet by _default_.
Lots of things require a restart when they are updated for the updates to take effect.
If you update apache or your GUI, you need to restart them.
Yes you can choose to not restart the stuff, but you are then NOT running the updated stuff.
Windows XP/7 doesn't force you to restart either. A restart is a simple way to ensure you are not running the out of date stuff.
[1] http://www.linux-mag.com/cache/7403/1.html
I'm not sure if that tech works with stuff like vmware. I always seem to need to do some stuff after kernel updates for vmware to work again (google for details)...
But that doesn't really sound like a Windows-specific problem to me.
That problem could happen with Linux too.
There's also lots of half-baked advice out there that tell users to delete certain stuff in windows\... while that frees up space, things like "windows update" can stop working.
And as we have seen, many users get infested with malware, and so updates cause problems (like BSODs).
Seems more like those lynch mobs.
Even if they aren't at fault (not saying they aren't) they have to do that apology thing etc.
It's a lose-lose.
If they deny it, they get in trouble - which driver is going to say he/she screwed up? If they say it's their fault, they get in trouble.
> For example, here in the Nordic countries, the words for "hour" begin with a "t", but we still use km/h to denote speed in kilometres per hour.
Over here the word for hour begins with j. But people understand both km/h and kph.
Lowercase b = bit. Uppercase B = byte.
In most cases it's better to buy pre-crimped certified cables and test them, than to crimp them and test them.
:).
How much money are you going to save? If you're charging $$$ for the project, crimping your own cables is a waste of time and resources. Unless of course your job is making cables
b (bits) and s (second) don't work in all languages either.
We spend lots of time complaining about people who confuse bps vs Bps, and milli with Mega, and there are the kibibyte vs kilobyte wars. And now b/s vs bps?
But you couldn't control exactly what was written with the Apple II drives.
:).
IIRC the spec was you could not expect to _reliably_ write a bit stream that has more than two zeroes in a row. In practice you could usually get away with three zeroes in a row. You had to do stuff like write sync streams e.g. 11111111001111111100111111110011111111001111111100. Which are basically 5 x 10 bits of 1111111100.
You could control the drive head too. You could even move the head in sub-track increments - e.g. half track. However, tracks have to be at least a full track apart to be reliably read/written.
All these is why I find the submitter's request rather strange. When you get down to the low level details, there is no such thing as exact.
Even if you are writing one bit on the entire platter with a huge magnet
So I'm curious on what he's really trying to achieve.
Just curious: are you satisfied with what you were getting for your money?
Would you prefer to pay less and have BBC make these cuts?
FWIW, the BBC is still one of the most respected media organizations in the world.
Unfortunately it's not very capitalistic or "free market"-ish to force everyone with a TV (or equivalent) to pay for the BBC. But if the alternative is more Rupert Murdoch and Friends...
If you drink a lot of water, maybe you should also add some salts.
Just drinking lots of water alone can be dangerous.
Not supposed to "light up" while using O2 :).
Some of those accelerator parts are made in the USA, some in Japan.
http://www.cbsnews.com/stories/2010/01/30/business/main6158199.shtml
Is Toyota's acceleration problem happening in significant numbers elsewhere in the world? Or is this limited to the USA?
No excuse in that case.
But what if all the attacker needs is for you to click on the link in the email?
e.g. overworked admin gets an email that appears to be from "That idiot ISP who keeps sending me announcements via http/pdf/msword links".
Admin clicks on link, gets pwned without knowing.
If the admin has html email enabled (just to be able to read instructions from bosses )the whole message could be a link too, so you could still accidentally click on the payload even if you know it is suspicious (e.g. misclick while trying to confirm the headers...).
The attackers may not need to resort to this yet since there are enough gullible people around. But looking at the number of drive-by exploits (whether for Windows, Linux or OSX) it doesn't give you much confidence that you won't ever get pwned.
Sorry about the jab back - I shouldn't, but just couldn't resist.
> Is there any actual study, however, which tested how "accessible" placebo is to atheists vs. believers? Much of the effectiveness, from what I've read, relies on the person believing that the (fake) treatment works.
My reasoning is that some believers can just pray to their God wherever they are. And even if their God doesn't actually help (because of nonexistence or other) they could still experience the placebo effect.
In contrast, atheists could try praying to the FSM, I currently doubt that's going to work as well.
HOWEVER I may be wrong since that has not been tested (to my knowledge). Perhaps the relevant parts of the brain could still work to produce the placebo effect when you pray to X even if you don't believe that X exists. That would be an interesting result. However I'm not sure if any reputable scientists are going to conduct such a study.
While dogs might actually believe in supernatural stuff, I doubt religion is required for the placebo effect, so one could not conclude that dogs are getting religion because placebos work on them.
To figure out whether dogs are getting religion one would have to conduct different experiments.
From the article at least one neighbour didn't mind.
Heck he even helped them:
Soon after the city complained about the yard, the Has placed wood chips on top of the dirt, with help from neighbor Dennis Cleek.
"It's their yard, it's not overgrown with weeds, it's not an eyesore," said Cleek, whose own yard boasts fruit trees. "We should be able to have our yards look the way we want them to."
And from the pic, it looks ok to me. As for wood-chips being a fire hazard, it's no big deal, before they start burning in a dangerous way due to some external cause, those wooden houses will probably be on fire first...
I propose astroturf then. Though that's still in violation (it's not living). But plastic grass is in the true spirit of LA right? :)