Please note that this is not intended to be used as your primary, everyday browser as it is not even to beta build quality. It will be updated as time permits to allow people to try out upcoming features of Chrome.
It's handy for trying it out without installing the full app and update service on Windows. Or for running Stable, Dev, Beta on the same machine with separate profiles without needing command line switches.
Yes, the Canary branch of Chrome uses a separate profile by default. Stable, Beta, and Dev all use the same default profile on Windows. You can run them with a command line option in the form --user-data-dir="C:\MyCustom\PathToMy\profile" to get them to use a specific profile, but they will still step on each other a bit in the registry within HKCU\Software\Google. The portable launchers for Google Chrome Portable's various channels will take care of any registry weirdness.
Thanks for mentioning us. I fixed a typo on the page as obviously the beta channel build is of beta quality :)
If you're interested, in addition to running Stable, Beta, and Dev side by side without impacting each other or local installs, you can create a duplicate of your GoogleChromePortable folder and have yet another copy there. Pro tip: If you name duplicates in the style of GoogleChromePortable_Copy_1, etc the PortableApps.com Platform will automatically update them all. If you name them something else like GoogleChromePortableOld, the updater will ignore them, so you can hang on to a specific version. You can grab older Chrome releases from our SourceForge project as well to test against previous versions alongside current, too.
The idealist in me cringes when he sees something like this. Why spoil something platonic, abstract, universal, like the web platform with something so accidential, petty, and bound to a specific culture and time, like payments? Not every thinkable society has money or will understand the concept of buying immaterial goods.
Then the realist reminds himself APIs have a churn time of a couple of years, whereas we will probably still be using money in a century.
With google bringing it's own adblocker to Google chrome (which is like calling up a casino for gambling addiction), I guess moving to open source chromium is the best idea now as long as Google is not dropping mysterious binary blobs to it (https://lwn.net/Articles/648392/)
> which is like calling up a casino for gambling addiction
A random aside - the gambling addiction helpline in the US includes an item in the IVR for making reservations at a casino (I believe it's the first item mentioned), or at least it did several years ago.
> which is like calling up a casino for gambling addiction
The analogy holds but I just want to point out that this is very common and all serious casino and poker businesses have staff working with terminating accounts for people requesting to be banned due to addiction. The same people are going over chat logs that get flagged due to mentioning of certain words like "addiction" and ban them even if they haven't requested it. If you tell a support agent that you're addicted you will be banned in an instant.
Is that because addiction is a recognized medical disability and thus the people would have a claim against the casino in case of losses? Or more for PR?
I can only speak for certain jurisdictions but apart from the reasons you mention it was a requirement from the Gambling Commission where I worked. In reality the casino found its ways to circumvent this for the whales though but that's a different topic.
They would probably publish some policy what is an acceptable ad and what isn't, and let the others self-regulate. For them, the ability to finger-point ("it's only your fault that you are not compliant with the blocking policy") is very important.
As a counterpoint, there's a This American Life story about how Harrah's/Caesar's casino hosts specifically marketed to people who clearly had gambling addictions:
Consider that there are plenty of ad blockers but no "unreasonable-ad blockers" (according to some policy). There's at least a niche here to be filled for people who only want some ads (the worst ones) to be blocked.
Compromise solutions do tend to be more popular with people in a position to benefit if a compromise can be reached.
Not sure if it's a Chrome update or macOS update as I did both at the same time, but since I updated to High Sierra I've started getting a lot of certificate errors in Chrome v59. I tried updating to the 60 Beta, but it is still occurring.
My company use a Cisco WSA to decrypt HTTPS traffic, and that cert is suddenly no longer trusted, and I seem to be unable to trust it. Despite it never being required to be trusted previously. Sites load fine in Safari though and off my corporate network.
Same issue at my job with McAfee proxies doing SSL interception.
Newest Chrome complains because it doesn't use the CN field anymore and now demands SAN (Subject Alternate Name) fields be present instead - but the certs generated by our proxy lack them.
This change has caused so much pain in our internal network too. Not because of intercepting proxies, but because of a ton of certs signed by an I terns CA.
I really wonder what kind of issue triggered them changing this and whether it was Wort the trouble. Stuff like this is why I can understand companies still being reluctant to allow browsers other than IE
i think it was because the rfc that says which attributes of a cert must be checked by the browser was deprecated 15 years ago and everybody ignored the update (that says that the san attribute must be present)
Gah! Is it blocking self-signed certs? I'd seen that in Canary, but it didn't show up in what I thought was the corresponding Chrome. At work, we use a lot of self-signed certs, and we have a hosted product that generates them too; figuring out a plan for handling that is no fun. The best workaround we found was using another browser.
>It just seems ridiculous to me to have 4 major releases in ‘17 alone
Hm, why is that ridiculous? Do you just mean in terms of having "breaking changes" every 3 months? I guess I've never heard of someone bit by this... they'd have to be going out of their way to force a downgrade to trigger breakage from this, no?
Major number is incremented when any user facing data or setting changes (ie, ~/.chrome might not be backwards compatible anymore 100% safely).
Minor number is incremented whenever.
The number that is most important is the third number, the four digit build number, which increments monotonically and never resets. That is a more meaningful number than the first two, IMO anyways.
It's really simple. The first number increments when a new release branch is cut off master, about every 6 weeks. The second is always zero. The third is a build number, which is an incrementing global.
There were exceptions in the early days when the team was still figuring things out. For example, release-branches were cut more sporadically, and the middle number was used to sorta-kinda represent smaller feature changes.
You might be thinking of release dates tied to a set of completed product features. Chrome doesn't work that way. The team ships on a regular schedule, holding back features that aren't ready for the public. Frequent releases help mitigate the urge for individual teams to ship low-quality stuff just because the next release train was a long and/or unpredictable time in the future.
If there is anyone from the Chrome team reading here: Please increase the number of suggestions while typing in the Url bar! In Firefox I can see 10 suggestion but in Chrome I can only see between 3 to 5 which is not enough.
This is especially annoying on Android where I prefer to do as little text input as possible and rather select urls from the suggestions.
A long time ago this value could be modified but that command line option was removed years ago.
Add a "1." before the version number in your head. It is all a matter of rescaling.
Of course, the real reason for the crazy version numbers back then was to 1. get people to update, 2. to look more recent than Firefox (which has since followed suit), and 3. to stop people from programming websites towards certain version numbers (because nobody has a clue anymore what version they are running). I don't like it, but I can follow their reasoning.
44 comments
[ 3.0 ms ] story [ 92.7 ms ] thread--
I was happy to learn Chrome is packaged by the PortableApps folks, simplifying side-by-side use of multiple versions with separate profiles.
https://portableapps.com/apps/internet/google_chrome_portabl...
Please note that this is not intended to be used as your primary, everyday browser as it is not even to beta build quality. It will be updated as time permits to allow people to try out upcoming features of Chrome.
source: https://news.ycombinator.com/item?id=14497989
It's handy for trying it out without installing the full app and update service on Windows. Or for running Stable, Dev, Beta on the same machine with separate profiles without needing command line switches.
If you're interested, in addition to running Stable, Beta, and Dev side by side without impacting each other or local installs, you can create a duplicate of your GoogleChromePortable folder and have yet another copy there. Pro tip: If you name duplicates in the style of GoogleChromePortable_Copy_1, etc the PortableApps.com Platform will automatically update them all. If you name them something else like GoogleChromePortableOld, the updater will ignore them, so you can hang on to a specific version. You can grab older Chrome releases from our SourceForge project as well to test against previous versions alongside current, too.
Press OK to pay $.50 to read this WSJ article!
Then the realist reminds himself APIs have a churn time of a couple of years, whereas we will probably still be using money in a century.
A random aside - the gambling addiction helpline in the US includes an item in the IVR for making reservations at a casino (I believe it's the first item mentioned), or at least it did several years ago.
The analogy holds but I just want to point out that this is very common and all serious casino and poker businesses have staff working with terminating accounts for people requesting to be banned due to addiction. The same people are going over chat logs that get flagged due to mentioning of certain words like "addiction" and ban them even if they haven't requested it. If you tell a support agent that you're addicted you will be banned in an instant.
it's going to start banning all ads except google ads.
They would probably publish some policy what is an acceptable ad and what isn't, and let the others self-regulate. For them, the ability to finger-point ("it's only your fault that you are not compliant with the blocking policy") is very important.
https://www.thisamericanlife.org/radio-archives/episode/466/...
Compromise solutions do tend to be more popular with people in a position to benefit if a compromise can be reached.
There are - such as https://adblockplus.org/acceptable-ads . This puts ad blockers in the ironic position of earning money via ad revenue...
My company use a Cisco WSA to decrypt HTTPS traffic, and that cert is suddenly no longer trusted, and I seem to be unable to trust it. Despite it never being required to be trusted previously. Sites load fine in Safari though and off my corporate network.
I really wonder what kind of issue triggered them changing this and whether it was Wort the trouble. Stuff like this is why I can understand companies still being reluctant to allow browsers other than IE
Could VP9 replace widevine?
It just seems ridiculous to me to have 4 major releases in ‘17 alone
Hm, why is that ridiculous? Do you just mean in terms of having "breaking changes" every 3 months? I guess I've never heard of someone bit by this... they'd have to be going out of their way to force a downgrade to trigger breakage from this, no?
Minor number is incremented whenever.
The number that is most important is the third number, the four digit build number, which increments monotonically and never resets. That is a more meaningful number than the first two, IMO anyways.
There were exceptions in the early days when the team was still figuring things out. For example, release-branches were cut more sporadically, and the middle number was used to sorta-kinda represent smaller feature changes.
You might be thinking of release dates tied to a set of completed product features. Chrome doesn't work that way. The team ships on a regular schedule, holding back features that aren't ready for the public. Frequent releases help mitigate the urge for individual teams to ship low-quality stuff just because the next release train was a long and/or unpredictable time in the future.
This is especially annoying on Android where I prefer to do as little text input as possible and rather select urls from the suggestions.
A long time ago this value could be modified but that command line option was removed years ago.
Of course, the real reason for the crazy version numbers back then was to 1. get people to update, 2. to look more recent than Firefox (which has since followed suit), and 3. to stop people from programming websites towards certain version numbers (because nobody has a clue anymore what version they are running). I don't like it, but I can follow their reasoning.
That still won't turn it into semver ( http://semver.org/ ), and 1.127 is not much better than 127.
Versioning can be not just random number to differentiate between two versions, but actually give useful information about changes to users and devs.