Mozilla had to remove XUL to make the browser properly multiprocess and multithreaded. If they hadn't done that, you would be telling that the browser is slow and insecure.
They will be able to choose whether they want to have UTC+0, UTC+1, UTC+2 or UTC+3
So the number of time zones won't change: currently you have +0, +1, +2 and summer time that adds one hour.
What really bothers me is that noon in some Spanish cities in the summer can be as late as 2:30pm (1:30pm in winter). They changed the time zone during WWII but didn't bother to change back.
If Spain chooses UTC+2, they will have sunrise in 9:37am in December.
From the actual bug report and commit in HG: it appears that this is only a new error page that appears instead of SEC_ERROR_UNKNOWN_ISSUER when Mozilla's update service detects a non-built-in cert.
So: this error will only appear if the current version displays unknown issuer error, and mozilla's update service detects that it has a MitM proxy.
Ok, I forgot about this, but it is an option that is shown if you click "choose what to share..." on new installations. There is an option with right click -> allow download that works when this happens.
How about the default google chrome's usage stats that they collect including Chrome User Experience Report that sends details about every page you visit.
You people talked about Firefox being slow before Firefox 56.
When they re-did parts of their engine in rust, removed XUL extensions support which prevented them from making the browser fully multithreaded, they made the browser significantly faster (comparable to chrome - faster in some tasks, slower in some other) - now you say that it is a clone of Chrome.
I don't like some of decisions Mozilla made in recent times, but still, Mozilla respects my privacy much more than Google ever will.
My laptop with skylake i5 runs 10 hours on battery with ubuntu 18.04 consumsles about 5-6 watts on idle and 6-8 watts when browsing... Windows consumes about 5 watts on idle so I don't see a difference.
I'm talking about latency increase over idle (caused by buffering). Are you sure you are testing your latency when transferring the data or when doing nothing?
On 0.5mbps upload connection, one 1500 byte packet takes 24ms to be transferred (with ADSL overhead it is about 30ms). That is the lowest latency increase over idle you can see when uploading data.
For downloads, your ISP has much more to do with your latency than your local network as they are the ones queuing packets to be sent over the slow (2.5Mbps) link, so the best you can do is make sure that TCP window scaling algorithm correctly sees that there is not enough bandwidth and stop sending packets before your ISP starts buffering them. The way to do this is to implement downstream shaping (using the same fq_codel/cake algorithm) and that will prevent the origin server from filling the buffers at your ISP DSLAM/CMTS.
Looks like https://www.bufferbloat.net/pr... would help a lot on your DSL line. It can be used on any openwrt compatible router https://wiki.openwrt.org/doc/h... . Also, a lot of Asus routers compatible with asuswrt-merlin have it https://github.com/RMerl/asusw...
It will help with games a lot - you will see about 30ms increased latency (time to send one full size packet at 0.5Mbps) when uploading over idle, and up to 50ms of increased latency when downloading (most of the time only about 5ms increased but at peaks it will go higher since your router are not at the correct location for doing QoS in the downstream).
My FTTH 200Mbps symmetric connection doesn't have good QoS on ISP's end, and without fq_codel/cake I see peak latency of about 100ms when downloading and 50ms when uploading. With it, I see almost no additional latency when uploading and about 1ms when downloading.
That's simply not true:
It took me 0 seconds to set up a Samsung Printer/Scanner combo (it worked just by plugging it in).
It took me 30 seconds to install a Kyocera printer - had to open printer settings (by clicking on a notification that popped up), click add printer, select the printer from the list and browse and select a (supplied by the manufacturer) PPD file.
It was no more difficult than opening a writer document using file open dialog.
The architectures removed are not used in mainstream computers - they are mainly used in signal processing or other embedded CPUs with low number of users.
Estonia has the least proprietary implementation - take a look at https://www.id.ee/?lang=en you can even compile the middleware from source, and on linux they interface with pcscd.
Linux can use PAE and address 32GB of RAM in 32bit. Each process can access up to 3GB of RAM, so you don't need 64bit if a single process doesn't use more than 3GB.
My Latitude E5470 (i5-6300u) uses 4.5W while idle at 50% brightness and achieves 12 hours and 20 minutes while idling. In windows 10, it uses 4.2W and achieves 13 hours in the same conditions.
My Latitude E5450 (i5-5300u) uses 5.2W while idle at 50% and achieves 10 hours. In windows, it uses 5.0W and achieves 10 hours 10 minutes.
My Vostro 5470 (i3-4010u) uses 4.5W while idle - 11 hours 30 minutes. In windows it uses 4.3W - 12 hours.
That's not a significant difference (5-7%), I can use the PCs for 7-8 hours when doing light browsing or coding.
My Latitude E5470 lasts 11h in Ubuntu 16.04 LTS and 11.5h in Windows 10 when not doing anything, with screen on - it consumes ~4W in linux and ~3.9W in windows 10.
I haven't had any issues with sleep - it works 100% of the time - I only reboot the system for kernel updates.
Intel and everyone else knows that restricted boot environments for personal computers (desktops and laptops) will be hugely profitable. Entertainment companies love it -- they can deploy a new kind of DRM that won't be defeated for years (see: PS3).
SecureBoot is not a DRM system (for now). If SecureBoot is on, the requirement is that the code executed before ExitBootServices() has to be signed. All code executed after that doesn't. So for example one can create a Boot Loader like EFILinux that will be signed and conform to the specification, and that can load unsigned kernels, and those unsigned kernels can contain any code. The kernel may emulate an EFI interface (like loaders for osx on BIOS), and load Windows kernel, patching it and then starting.
Or, you on PCs that have it turned off, you can create your own EFI application that will load instead of windows's boot loader that will override the GetVariables() functions, so that if the windows kernel queries it, it will return that the SecureBoot is On. It can also patch the kernel itself in memory before starting it.
Slashdot Mirror
Mozilla had to remove XUL to make the browser properly multiprocess and multithreaded. If they hadn't done that, you would be telling that the browser is slow and insecure.
All EU countries will have to stop using DST.
They will be able to choose whether they want to have UTC+0, UTC+1, UTC+2 or UTC+3
So the number of time zones won't change: currently you have +0, +1, +2 and summer time that adds one hour.
What really bothers me is that noon in some Spanish cities in the summer can be as late as 2:30pm (1:30pm in winter). They changed the time zone during WWII but didn't bother to change back.
If Spain chooses UTC+2, they will have sunrise in 9:37am in December.
It won't: see my comment.
From the actual bug report and commit in HG: it appears that this is only a new error page that appears instead of SEC_ERROR_UNKNOWN_ISSUER when Mozilla's update service detects a non-built-in cert.
So: this error will only appear if the current version displays unknown issuer error, and mozilla's update service detects that it has a MitM proxy.
I would suggest a great alternative I used for 8 years, but they got bought by GoDaddy :(
Ok, I forgot about this, but it is an option that is shown if you click "choose what to share..." on new installations. There is an option with right click -> allow download that works when this happens.
How about the default google chrome's usage stats that they collect including Chrome User Experience Report that sends details about every page you visit.
You people talked about Firefox being slow before Firefox 56.
When they re-did parts of their engine in rust, removed XUL extensions support which prevented them from making the browser fully multithreaded, they made the browser significantly faster (comparable to chrome - faster in some tasks, slower in some other) - now you say that it is a clone of Chrome.
I don't like some of decisions Mozilla made in recent times, but still, Mozilla respects my privacy much more than Google ever will.
My laptop with skylake i5 runs 10 hours on battery with ubuntu 18.04 consumsles about 5-6 watts on idle and 6-8 watts when browsing... Windows consumes about 5 watts on idle so I don't see a difference.
Take a look at: https://github.com/gnif/Lookin...
Take a look at https://docs.microsoft.com/en-... section 10.2.1
I'm talking about latency increase over idle (caused by buffering). Are you sure you are testing your latency when transferring the data or when doing nothing? On 0.5mbps upload connection, one 1500 byte packet takes 24ms to be transferred (with ADSL overhead it is about 30ms). That is the lowest latency increase over idle you can see when uploading data. For downloads, your ISP has much more to do with your latency than your local network as they are the ones queuing packets to be sent over the slow (2.5Mbps) link, so the best you can do is make sure that TCP window scaling algorithm correctly sees that there is not enough bandwidth and stop sending packets before your ISP starts buffering them. The way to do this is to implement downstream shaping (using the same fq_codel/cake algorithm) and that will prevent the origin server from filling the buffers at your ISP DSLAM/CMTS.
Looks like https://www.bufferbloat.net/pr... would help a lot on your DSL line. It can be used on any openwrt compatible router https://wiki.openwrt.org/doc/h... . Also, a lot of Asus routers compatible with asuswrt-merlin have it https://github.com/RMerl/asusw... It will help with games a lot - you will see about 30ms increased latency (time to send one full size packet at 0.5Mbps) when uploading over idle, and up to 50ms of increased latency when downloading (most of the time only about 5ms increased but at peaks it will go higher since your router are not at the correct location for doing QoS in the downstream). My FTTH 200Mbps symmetric connection doesn't have good QoS on ISP's end, and without fq_codel/cake I see peak latency of about 100ms when downloading and 50ms when uploading. With it, I see almost no additional latency when uploading and about 1ms when downloading.
That's simply not true: It took me 0 seconds to set up a Samsung Printer/Scanner combo (it worked just by plugging it in). It took me 30 seconds to install a Kyocera printer - had to open printer settings (by clicking on a notification that popped up), click add printer, select the printer from the list and browse and select a (supplied by the manufacturer) PPD file. It was no more difficult than opening a writer document using file open dialog.
The architectures removed are not used in mainstream computers - they are mainly used in signal processing or other embedded CPUs with low number of users.
I found the source for their .deb packages here https://installer.id.ee/media/... so it shouldn't be complicated to compile it on other linux distros.
Estonia has the least proprietary implementation - take a look at https://www.id.ee/?lang=en you can even compile the middleware from source, and on linux they interface with pcscd.
Linux can use PAE and address 32GB of RAM in 32bit. Each process can access up to 3GB of RAM, so you don't need 64bit if a single process doesn't use more than 3GB.
My Latitude E5470 (i5-6300u) uses 4.5W while idle at 50% brightness and achieves 12 hours and 20 minutes while idling. In windows 10, it uses 4.2W and achieves 13 hours in the same conditions. My Latitude E5450 (i5-5300u) uses 5.2W while idle at 50% and achieves 10 hours. In windows, it uses 5.0W and achieves 10 hours 10 minutes. My Vostro 5470 (i3-4010u) uses 4.5W while idle - 11 hours 30 minutes. In windows it uses 4.3W - 12 hours. That's not a significant difference (5-7%), I can use the PCs for 7-8 hours when doing light browsing or coding.
My Latitude E5470 lasts 11h in Ubuntu 16.04 LTS and 11.5h in Windows 10 when not doing anything, with screen on - it consumes ~4W in linux and ~3.9W in windows 10. I haven't had any issues with sleep - it works 100% of the time - I only reboot the system for kernel updates.
Not true, the version that supports XP is Firefox 52 ESR which still supports Java and will until mid 2018.
Isn't it 2500 HDDs (assuming 6TB each) a year?
You could try adding ppa:mc3man/trusty-media. It worked for me.
I don't think it works on anything but webkit. I've tried in in opera mobile for android, and in mobile firefox.
Intel and everyone else knows that restricted boot environments for personal computers (desktops and laptops) will be hugely profitable. Entertainment companies love it -- they can deploy a new kind of DRM that won't be defeated for years (see: PS3).
SecureBoot is not a DRM system (for now). If SecureBoot is on, the requirement is that the code executed before ExitBootServices() has to be signed. All code executed after that doesn't. So for example one can create a Boot Loader like EFILinux that will be signed and conform to the specification, and that can load unsigned kernels, and those unsigned kernels can contain any code. The kernel may emulate an EFI interface (like loaders for osx on BIOS), and load Windows kernel, patching it and then starting.
Or, you on PCs that have it turned off, you can create your own EFI application that will load instead of windows's boot loader that will override the GetVariables() functions, so that if the windows kernel queries it, it will return that the SecureBoot is On. It can also patch the kernel itself in memory before starting it.