Interesting... I have to say I wasn't overly impressed by the bits of wasm implementation I saw when it was first announced. Sounds like they have gotten further along but also made a few mistakes along the way. Like the JVM not having unsigned ints, little things like that can have really icky consequences when the rubber hits the road.
I wonder if the control flow prohibition is cultural or technical baggage from the "I have an axe so every problem fits in a tree structure" DOM shortcomings.
Still using ESR. Took qupzilla for a test drive for a while because it still had the separate search engine toolbar element and wasn't going to drop ALSA support AFIACT, but it's still just a bit too broken for use on some of the sites I visit.
It's "I need to schedule a meeting with these people, when are most of them available, and can I book a conference room too."
I have no doubt that's indispensable for some environments. In mine, this is how a rational individual would go about organizing a meeting, if a meeting is even merited (most matters are email-resolvable if your staff is sufficiently literate.)
1) Think who the least likely person to have time to do it and also be prepared for the meeting is, call or visit their office, and ask when they'd be both prepared and have time. Because that won;t match when they are "free" on their calander.
2) Rinse and repeat for a few more most contended people.
3) Check room availability.
4) Propose a meeting time and alternative to the whole group you want to attend.
5) Deal with any problems with that.
This is how it works with a calendaring system where I work:
1) Spam everyone with a meeting announcement at some random time chosen off their free time. 2) Half the people check their off-line calanders and decide how they feel about the time and confirm. 3) One guy has a problem with the time because of something external to his "availability" timeline. 4) Propose a new time, which at least one of the intended participants fails to notice is a time change 5) Eventually after however many iterations get everyone on the same page. 6) Decide arbitrarily to change the meeting room, somehow screw up the change so there is no announcement email about the change. 7) Wonder where Hal is at the meeting. He's looking for the meeting in the old room. 8) Wonder where Sally is. She glossed over the meeting announcement and didn't notice the time was different from the time you normally hold such meetings because the originator failed to draw special attention to this in the text of the announcement. 9) Wonder where Ned is. His Windows desktop went to sleep while he was deep in concentration debugging some code using his Linux system, and he didn't get the reminder.
Take a moment to realize that what is being referred to here is not the developed world, where population tends to stabilize, even though footprint increases. It's the post-agrarian/tribal cultures that have large families and isolationism deeply ingrained in their traditions. It takes slow, patient work to untangle the Gordian knot of religion/mores/entrenchment. Also when you try to change someone's culture it takes a whole ass-ton of humility to not get run out on a rail. So, the bible-thumping anti-birth-control missionaries of the last several centuries were plainly a highly suboptimal approach.
I agree, calendaring is what's missing in the FOSS world.
I both agree and disagree. Calendaring integration (and nowadays chat/voice integration) is what the PHBs cannot live without.
It's also one of the most unnecessary and counterproductive applications for the majority of the workforce. You know, the ones that don't waste all their time in meetings, know how to communicate well by email, and actually have work they need to concentrate on. To them, calendaring is a way for you to make them reorganize their day 5 times over as PHB moves their stupid little meeting block around to make room for some other stupid little meeting block.
(Not to mention Exchange's calendaring is pretty messed up, what with meeting emails that just disappear automatically after the first time you open them so you have to go dig in your calendar app for them, meeting changes that only sometimes get you an email to say that your meeting times changed, random emails scheduling you for half-edited meetings that the user of the app didn't mean to send or tried to cancel, meeting "resources" with permissions on them that nobody seems to be aware prevent a good number of people from actually interacting with the calendar entry, etc. etc. etc.)
There are some people who legitimately do a lot of work that legitimately involves lots of meetings but who do not rise to the level of needing a secretary to organize their day for them. For them calandering is probably useful. For the rest of us with other less social duties to perform, the whole thing is a midsized time vampire we are expected to keep an eye on while we are elbow deep in important work.
Larger key travel helps to not accidentally type a letter if you scuff an adjoining key. When that happens you can sort of fall off the edge of the scuffed key and manage to get the one you wanted on the side anyway. In the case of a wide-spaced chiclet, you are just hitting the laptop frame. Also the chiclet tops are not the same as a full size key... full size keys are cupped, and the tactile feedback from the cup/corner shape helps you align better without looking.
I've pretty much given up hope at this point, and when not using a full-size, I revert to hunt-and-peck-really-hard. Kills my CPS, but at least I manage to get every character to actually register that way. Don't understand how such an active consumer market can continually produce worse and worse product over time. You'd think some "invisible hand" would eventually make its appearance on capitalism's home turf.
I was kinda hoping the scissor hinges would fix chiclets always missing keystrokes... haven't had a chance to try even that yet... but it sounds like they found yet more ways to make it suck. If something like the Thinkpad 25 is still around when I come due to a refresh, I'll definitely give it a look... I don't need huge performance out of this thing, just so long as it has an SSD for faster compiles, the CPU doesn't matter much. But... I'm getting on in years, so a bigger wide format screen definitely helps.
In addition to modern IQ scores being back-fitted to a bell curve (one could argue this fudging serves very llittle useful purpose or for or against the evidence that a bell curve is the correct distribution to use) there is also the definition of "average":
a number expressing the central or typical value in a set of data, in particular the mode, median, or (most commonly) the mean, which is calculated by dividing the sum of the values in the set by their number.
...it can mean any of those when used in common discourse.
All of this is beside the point. Which is... who the heck thought it was a good idea to link a vibrator to a phone app in the first place. What about a vibrator screams: "I really need to be linked to an app on an internet-connected device?" What sensible individual would make a decision to proceed with such a product?
Yeah, when they make a stylus that emulates bumps in the paper that I can feel, maybe I'll give it a try.
But, paraphrasing some football coach recently, you know what you can do with a pad of paper you can;t with a tablet? Throw it on the ground and stomp on it in rage.
Also it the contents don't tend to accidentally end up on the cloud when you don't want them to.
Real paper has a frictional quality to it you'll never get on gorilla glass.
Anyway if any of these e-ink gadgets are cheap enough It'd be a great alternative to putting my cell phone under my mouse and making it vibrate every few minutes to keep the lock screen from kicking in on my corporate MS box while I'm busy ignoring my email and working on my linux box. Just a pad with a shifting pattern.
Well, most places haven't gotten to quite this level yet, when combined with a EAPOL EAP-TLS/dot1x you register your MAC address, you get a client cert containing that MAC address, and the switch will not let you on using a different MAC address than one you have registered. Presumably you don't allow double-registrations, of course. (Further past that you can close the last of the wired MITM vectors with MACSec but that requires rather new switches still.)
But even with (yes, easily spoofable) MAC addresses there are several benefits. First, only allowing registered MAC addresses on the network prevents people flooding the network ARP and bridging tables with lots of fake MAC addresses or exhausting your DHCP pools, so you close some DOS and flooding attack vectors that can completely wreck the network or allow messing with traffic. Second, most people are not spoofing, and knowing who owns that machine that got infected shortens incident response time. It's also the most convenient thing to use as a machine identifier, lacking the GUID you get in managed environments.
Cancer's starting a downtrend as well. Next up is either COPD or age related neurologicals. Though the pill epidemic is certainly taking its toll.
I don't know why so many people react so badly to news of anti-senescence research. Maybe they are afraid if they live too long all their bullshit will catch up to them?
Also, I'm puzzled by the moral perspective comment
I meant the good of environmental protection is in the same bag as the rest of the heavy handed policies. Also, though the end result on the environment is good, the means do matter, and in the past some environmental reforms have included death penalties and probably (not knowing of an example) corruption-tainted company takeovers.
Sentient life is incredibly improbable, therefore sentient life ends up observing an incredibly improbable universe around it, because in all the other possible outcomes, there's nobody there to observe it. So maybe it's just that we are in the matter corner of a universe that just lucked out, and the corresponding antimatter part is somewhere outside the observable portion of the universe.
China's main political push right now is preventing societal instability... the government is doing so with quite a heavy hand when it comes to politics with severe harassment of journalists, lawyers, and activists. One thing that has been a major source of unrest in the past is riots due to insufferable environmental conditions, so this would logically fit on the list of things to do towards quelling unrest. So it's part of a mixed bag from a moral perspective.
Despite a better commitment to renewables than the U.S has, China is playing catch-up and has a long way to close the gap in environmental protections... but hey, since we are currently backsliding at the federal level that makes catching up easier for them at least, I guess...
I think they are talking about elective/cosmetic surgery where there is no immediate pain or risk to health aren't they?
Were it the health care system I had to go to, I would never assume such a thing. There are plenty of abstinence-only tobacco zealots available to formulate policies that have no solid scientific grounding in either overall patient welfare, or in legitimate health system financial concerns. Plenty of fatty-punchers, too. And, if previous comments have any merit, plenty of people willing to take extreme positions opposite of their actual agenda in order to discredit the system as a whole in an attempt to undermine it.
Yeah good luck with that. Sometime in my 30s I stopped being able to make anything even resembling my own signature. I can't for the life of me handwrite two signatures that look anything alike right under each other. And I'm from a generation that had hours and hours of penmanship drills in grammar school. I can't even imagine how diverse signatures have gotten now that handwriting is something of a weekly event rather than several times a day.
FWIW, "move fast and break things" developers don't generally last very long in college environments. Software dev runs at a snails pace because everyone is actually using the software for important things, and when it breaks, there is hell to pay from the users... and the users are only a short walk away from you.
One MAC address per port is a bit on the extreme side for a residential user hospitality port... probably they have just not yet bought equipment capable of multi-client wired MAB/dot1x thus the one-MAC-address limitation. Registration of your MAC addresses is absolutely essential for security as, with wired dot1x and a cert bearing your registration, you can shut down a huge number of amateur-skill attacks by doing that (and if you haven't gotten all the way to wired dot1x yet and are still using MAB, you can at least raise the bar a little and also prevent those amateur attacks which can take down the whole network.)
However "LAN gaming" is kind of ridiculous to expect any enterprise LAN to support... these are segmented networks out of necessity. The "One Big LAN" model just collapses under its own weight for technical reasons at college student body scales. You have to switch to NBMA models (preferred in the case of WiFi to limit RF chatter) or segment or the whole network or it will just crash all the time due to BC storms. That isn't a matter of a bugs in the gear, it's just a natural consequence of the ethernet standard.
So who is on your "LAN" would be pretty arbitrary. It could be just your dorm floor or it could be only the people in your same class year or academic major or whatnot. On wireless, turning off broadcast/multicast is common best practice, and that would shut down any LAN game that relies on broadcast/multicast for discovering the participants, or during gameplay. Sometimes you have the staff and funds to administer a filtered/cross-vlan tool providing SDPish services to re-enable some of the living-room crap people want like sharing to their AppleTVs, sometimes you don't, but I don't know of any product of that stripe that has prebuilt profiles for the huge assortment of ad-hoc game protocols that must be out there.
Forbidding local air-gapped networks is just beyond the pale. That college IT department needs to be put into its place. And, perhaps, resourced to configure STP and other loopguard protections correctly, since they probably did that just because they were afraid of network loops. They are just inviting disaster since some person within WfI range and out of their policy jurisdiction could fire up an alternative service, and given the oppressive restrictions, people will jump on it whether it is secure or not. You have to keep the network functional enough for the users to want to stay on it, or your security is threatened.
Pretty much... I've had to evaluate security solutions hailing from the corporate sector for application in.edu, and I have to say so many of them put a disturbing amount of trust on their abilty to lock down the client OSes. Now this makes them pretty much useless in an environment where joining the majority of devices on the network to a domain or MDM is just plainly not an option (the users won't stand for it and even if they did, we have continuing ed users with conflicting configs on the work laptops from other companies which they bring to class). But even if we were able to do so, you should pretty much never trust client machines, even if you've gone all in on the even-with-TPM-won't-even-boot-BIOS-unless-connected-to-a-cloud-verification-service crap. You have to harden the infrastructure as if it were an internet-facing service, (while still doing what you can on the network layer to restrict access and at the OS layer to keep machines updated.)
Two words: supplemental policy. The math isn't that hard, try using your brain.
However, one does have to ask whether this is a wise choice based on the evidence. If the patient is in pain, for example, forcing that patient into the traumatic experience of withdrawal may be contraindicated, and if the surgery has preventative merits, patients may delay to avoid that experience and end up burdening the system even more once the need progresses to an urgent state.
Slashdot Mirror
Interesting... I have to say I wasn't overly impressed by the bits of wasm implementation I saw when it was first announced. Sounds like they have gotten further along but also made a few mistakes along the way. Like the JVM not having unsigned ints, little things like that can have really icky consequences when the rubber hits the road.
I wonder if the control flow prohibition is cultural or technical baggage from the "I have an axe so every problem fits in a tree structure" DOM shortcomings.
Still using ESR. Took qupzilla for a test drive for a while because it still had the separate search engine toolbar element and wasn't going to drop ALSA support AFIACT, but it's still just a bit too broken for use on some of the sites I visit.
It's "I need to schedule a meeting with these people, when are most of them available, and can I book a conference room too."
I have no doubt that's indispensable for some environments. In mine, this is how a rational individual would go about organizing a meeting, if a meeting is even merited (most matters are email-resolvable if your staff is sufficiently literate.)
1) Think who the least likely person to have time to do it and also be prepared for the meeting is, call or visit their office, and ask when they'd be both prepared and have time. Because that won;t match when they are "free" on their calander.
2) Rinse and repeat for a few more most contended people.
3) Check room availability.
4) Propose a meeting time and alternative to the whole group you want to attend.
5) Deal with any problems with that.
This is how it works with a calendaring system where I work:
1) Spam everyone with a meeting announcement at some random time chosen off their free time.
2) Half the people check their off-line calanders and decide how they feel about the time and confirm.
3) One guy has a problem with the time because of something external to his "availability" timeline.
4) Propose a new time, which at least one of the intended participants fails to notice is a time change
5) Eventually after however many iterations get everyone on the same page.
6) Decide arbitrarily to change the meeting room, somehow screw up the change so there is no announcement email about the change.
7) Wonder where Hal is at the meeting. He's looking for the meeting in the old room.
8) Wonder where Sally is. She glossed over the meeting announcement and didn't notice the time was different from the time you normally hold such meetings because the originator failed to draw special attention to this in the text of the announcement.
9) Wonder where Ned is. His Windows desktop went to sleep while he was deep in concentration debugging some code using his Linux system, and he didn't get the reminder.
Take a moment to realize that what is being referred to here is not the developed world, where population tends to stabilize, even though footprint increases. It's the post-agrarian/tribal cultures that have large families and isolationism deeply ingrained in their traditions. It takes slow, patient work to untangle the Gordian knot of religion/mores/entrenchment. Also when you try to change someone's culture it takes a whole ass-ton of humility to not get run out on a rail. So, the bible-thumping anti-birth-control missionaries of the last several centuries were plainly a highly suboptimal approach.
feels like you're stuck in the 90s
What, is it responsive, discoverable, and well layed out? Golden age of software, there.
I agree, calendaring is what's missing in the FOSS world.
I both agree and disagree. Calendaring integration (and nowadays chat/voice integration) is what the PHBs cannot live without.
It's also one of the most unnecessary and counterproductive applications for the majority of the workforce. You know, the ones that don't waste all their time in meetings, know how to communicate well by email, and actually have work they need to concentrate on. To them, calendaring is a way for you to make them reorganize their day 5 times over as PHB moves their stupid little meeting block around to make room for some other stupid little meeting block.
(Not to mention Exchange's calendaring is pretty messed up, what with meeting emails that just disappear automatically after the first time you open them so you have to go dig in your calendar app for them, meeting changes that only sometimes get you an email to say that your meeting times changed, random emails scheduling you for half-edited meetings that the user of the app didn't mean to send or tried to cancel, meeting "resources" with permissions on them that nobody seems to be aware prevent a good number of people from actually interacting with the calendar entry, etc. etc. etc.)
There are some people who legitimately do a lot of work that legitimately involves lots of meetings but who do not rise to the level of needing a secretary to organize their day for them. For them calandering is probably useful. For the rest of us with other less social duties to perform, the whole thing is a midsized time vampire we are expected to keep an eye on while we are elbow deep in important work.
Larger key travel helps to not accidentally type a letter if you scuff an adjoining key. When that happens you can sort of fall off the edge of the scuffed key and manage to get the one you wanted on the side anyway. In the case of a wide-spaced chiclet, you are just hitting the laptop frame. Also the chiclet tops are not the same as a full size key... full size keys are cupped, and the tactile feedback from the cup/corner shape helps you align better without looking.
I've pretty much given up hope at this point, and when not using a full-size, I revert to hunt-and-peck-really-hard. Kills my CPS, but at least I manage to get every character to actually register that way. Don't understand how such an active consumer market can continually produce worse and worse product over time. You'd think some "invisible hand" would eventually make its appearance on capitalism's home turf.
I was kinda hoping the scissor hinges would fix chiclets always missing keystrokes... haven't had a chance to try even that yet... but it sounds like they found yet more ways to make it suck. If something like the Thinkpad 25 is still around when I come due to a refresh, I'll definitely give it a look... I don't need huge performance out of this thing, just so long as it has an SSD for faster compiles, the CPU doesn't matter much. But... I'm getting on in years, so a bigger wide format screen definitely helps.
In addition to modern IQ scores being back-fitted to a bell curve (one could argue this fudging serves very llittle useful purpose or for or against the evidence that a bell curve is the correct distribution to use) there is also the definition of "average":
a number expressing the central or typical value in a set of data, in particular the mode, median, or (most commonly) the mean, which is calculated by dividing the sum of the values in the set by their number.
...it can mean any of those when used in common discourse.
All of this is beside the point. Which is... who the heck thought it was a good idea to link a vibrator to a phone app in the first place. What about a vibrator screams: "I really need to be linked to an app on an internet-connected device?" What sensible individual would make a decision to proceed with such a product?
Yeah, when they make a stylus that emulates bumps in the paper that I can feel, maybe I'll give it a try.
But, paraphrasing some football coach recently, you know what you can do with a pad of paper you can;t with a tablet? Throw it on the ground and stomp on it in rage.
Also it the contents don't tend to accidentally end up on the cloud when you don't want them to.
Real paper has a frictional quality to it you'll never get on gorilla glass.
Anyway if any of these e-ink gadgets are cheap enough It'd be a great alternative to putting my cell phone under my mouse and making it vibrate every few minutes to keep the lock screen from kicking in on my corporate MS box while I'm busy ignoring my email and working on my linux box. Just a pad with a shifting pattern.
I was thinking "refrigerator magnet"
Well, most places haven't gotten to quite this level yet, when combined with a EAPOL EAP-TLS/dot1x you register your MAC address, you get a client cert containing that MAC address, and the switch will not let you on using a different MAC address than one you have registered. Presumably you don't allow double-registrations, of course. (Further past that you can close the last of the wired MITM vectors with MACSec but that requires rather new switches still.)
But even with (yes, easily spoofable) MAC addresses there are several benefits. First, only allowing registered MAC addresses on the network prevents people flooding the network ARP and bridging tables with lots of fake MAC addresses or exhausting your DHCP pools, so you close some DOS and flooding attack vectors that can completely wreck the network or allow messing with traffic. Second, most people are not spoofing, and knowing who owns that machine that got infected shortens incident response time. It's also the most convenient thing to use as a machine identifier, lacking the GUID you get in managed environments.
Cancer's starting a downtrend as well. Next up is either COPD or age related neurologicals. Though the pill epidemic is certainly taking its toll.
I don't know why so many people react so badly to news of anti-senescence research. Maybe they are afraid if they live too long all their bullshit will catch up to them?
Whoosh... read up a few headlines.
Also, I'm puzzled by the moral perspective comment
I meant the good of environmental protection is in the same bag as the rest of the heavy handed policies. Also, though the end result on the environment is good, the means do matter, and in the past some environmental reforms have included death penalties and probably (not knowing of an example) corruption-tainted company takeovers.
It's even possible at small scale given enough time. Lots and lots of time. An example being the Loschmidt paradox and the Fluctuation Theorem.
Sentient life is incredibly improbable, therefore sentient life ends up observing an incredibly improbable universe around it, because in all the other possible outcomes, there's nobody there to observe it. So maybe it's just that we are in the matter corner of a universe that just lucked out, and the corresponding antimatter part is somewhere outside the observable portion of the universe.
China's main political push right now is preventing societal instability... the government is doing so with quite a heavy hand when it comes to politics with severe harassment of journalists, lawyers, and activists. One thing that has been a major source of unrest in the past is riots due to insufferable environmental conditions, so this would logically fit on the list of things to do towards quelling unrest. So it's part of a mixed bag from a moral perspective.
Despite a better commitment to renewables than the U.S has, China is playing catch-up and has a long way to close the gap in environmental protections... but hey, since we are currently backsliding at the federal level that makes catching up easier for them at least, I guess...
Apparently spell-checked garbage at least... compared eith the actual headlines.
I think they are talking about elective/cosmetic surgery where there is no immediate pain or risk to health aren't they?
Were it the health care system I had to go to, I would never assume such a thing. There are plenty of abstinence-only tobacco zealots available to formulate policies that have no solid scientific grounding in either overall patient welfare, or in legitimate health system financial concerns. Plenty of fatty-punchers, too. And, if previous comments have any merit, plenty of people willing to take extreme positions opposite of their actual agenda in order to discredit the system as a whole in an attempt to undermine it.
Yeah good luck with that. Sometime in my 30s I stopped being able to make anything even resembling my own signature. I can't for the life of me handwrite two signatures that look anything alike right under each other. And I'm from a generation that had hours and hours of penmanship drills in grammar school. I can't even imagine how diverse signatures have gotten now that handwriting is something of a weekly event rather than several times a day.
FWIW, "move fast and break things" developers don't generally last very long in college environments. Software dev runs at a snails pace because everyone is actually using the software for important things, and when it breaks, there is hell to pay from the users... and the users are only a short walk away from you.
One MAC address per port is a bit on the extreme side for a residential user hospitality port... probably they have just not yet bought equipment capable of multi-client wired MAB/dot1x thus the one-MAC-address limitation. Registration of your MAC addresses is absolutely essential for security as, with wired dot1x and a cert bearing your registration, you can shut down a huge number of amateur-skill attacks by doing that (and if you haven't gotten all the way to wired dot1x yet and are still using MAB, you can at least raise the bar a little and also prevent those amateur attacks which can take down the whole network.)
However "LAN gaming" is kind of ridiculous to expect any enterprise LAN to support... these are segmented networks out of necessity. The "One Big LAN" model just collapses under its own weight for technical reasons at college student body scales. You have to switch to NBMA models (preferred in the case of WiFi to limit RF chatter) or segment or the whole network or it will just crash all the time due to BC storms. That isn't a matter of a bugs in the gear, it's just a natural consequence of the ethernet standard.
So who is on your "LAN" would be pretty arbitrary. It could be just your dorm floor or it could be only the people in your same class year or academic major or whatnot. On wireless, turning off broadcast/multicast is common best practice, and that would shut down any LAN game that relies on broadcast/multicast for discovering the participants, or during gameplay. Sometimes you have the staff and funds to administer a filtered/cross-vlan tool providing SDPish services to re-enable some of the living-room crap people want like sharing to their AppleTVs, sometimes you don't, but I don't know of any product of that stripe that has prebuilt profiles for the huge assortment of ad-hoc game protocols that must be out there.
Forbidding local air-gapped networks is just beyond the pale. That college IT department needs to be put into its place. And, perhaps, resourced to configure STP and other loopguard protections correctly, since they probably did that just because they were afraid of network loops. They are just inviting disaster since some person within WfI range and out of their policy jurisdiction could fire up an alternative service, and given the oppressive restrictions, people will jump on it whether it is secure or not. You have to keep the network functional enough for the users to want to stay on it, or your security is threatened.
Pretty much... I've had to evaluate security solutions hailing from the corporate sector for application in .edu, and I have to say so many of them put a disturbing amount of trust on their abilty to lock down the client OSes. Now this makes them pretty much useless in an environment where joining the majority of devices on the network to a domain or MDM is just plainly not an option (the users won't stand for it and even if they did, we have continuing ed users with conflicting configs on the work laptops from other companies which they bring to class). But even if we were able to do so, you should pretty much never trust client machines, even if you've gone all in on the even-with-TPM-won't-even-boot-BIOS-unless-connected-to-a-cloud-verification-service crap. You have to harden the infrastructure as if it were an internet-facing service, (while still doing what you can on the network layer to restrict access and at the OS layer to keep machines updated.)
Two words: supplemental policy. The math isn't that hard, try using your brain.
However, one does have to ask whether this is a wise choice based on the evidence. If the patient is in pain, for example, forcing that patient into the traumatic experience of withdrawal may be contraindicated, and if the surgery has preventative merits, patients may delay to avoid that experience and end up burdening the system even more once the need progresses to an urgent state.