The thing that impressed me at first with Vivaldi was the snappier loading of pages.

The one thing I am missing with Vivaldi is being able to visit DoD Smartcard/CAC enabled sites. I have the necessary drivers and libraries installed. On Firefox I can go into Security Devices and load the Coolkey module (actually, it was loaded automatically once everything was installed and running), and it brings up my Smartcard credentials when I go to those sites. In Vivaldi, I just get a ERR_BAD_SSL_CLIENT_AUTH_CERT error. I've found a few Smartcard Chrome extensions, but they seem to be intended for ChromeOS and don't have an effect here. I don't suppose anyone has insight on this particular use case?

According to this, you need to install mozilla-nss-tools. But it's been a few years since I attempted to get my CAC working on Linux. Maybe I'll give it a go again soon... I do have some CAC readers floating around here.

Hm, no luck. I'm getting the bad database error when running the initial modutil command. I will have to look into it more later. Thanks for the lead, though.

Nevermind, I should have read more carefully. I was not in $HOME when I ran the command. Now it works. Thanks bassmadrigal!

Looks like the Slackbuilds site has the widevine now as a download. Thanks Alex

Firefox was annoying me after many years of being quite happy with it so I thought I'd give Vivaldi another try.

Wow, vast improvement over the last time I tried it. Haven't fired up Firefox since.

Thanks to all involved for a really good alternative.

When I brought it up, of course, I couldn't find one of these, but just now stumbled across this one another board while running Vivaldi. Notice it is reporting Vivaldi as "Chrome."

Since I am back from vacation it looks like I have a few things in this thread I should reply to and/or address but it is late now in Oslo, so that will have to be tomorrow.

In the mean time I followed slalik's lead and updated latest-vivaldi.sh to handle Widevine automatically as well.

[Edit]: I just (@Thu Jan 3 20:02 CET 2019) tweaked the script again slightly as I realised that my method could fail in some minor cases.

Thanks for this. I see you took a slightly different approach than me (which is absolutely fine!).

Yes, because it is running Chromium base, some sites see it as Chrome/Chromium. Others will see it as Vivaldi.

Sure you can do that if you want but my fonts already look pretty good and I don't do that. Instead I used @dugan's awesome blog post about font configuration on Slackware and made some tweaks, including using a self compiled freetype, with the patch applied to enable subpixel rendering in FreeType. I changed a few other things as well IIRC, though I did not keep my own notes as I figured I would just go back to dugan's page if I needed to setup another machine.

Well you can use tabs without a tab bar. Enable the tab cycler and you will see them on Ctrl-Tab. Or search though open tabs with Quick Commands or by opening the window panel.

Yeah but I don't think that is what he means. He means, how do websites gathering statics count Vivaldi. Do they count it as Vivaldi or as Chrome (and clones)? That is why he later said this:

Lysander666 also misunderstood him but has the right idea.

Indeed, the answer depends entirely on the website capturing statistics and how much effort the take to parse the User Agent. In addition we lie about our user Agent to a bunch of sites, e.g. we claim to be Chrome to most Google sites. If we did not they give us a different (crapper or non-working) version of their page or tell us we are unsupported. We also lie to whatsapp, netflix and even facebook (for Vivaldi on macOS only) for similar reasons. There are mroe sites but I can't be bothered to look them all up. ;)

+1 to that!

While there are a few things we have tweaked that may be causing a difference in speed for the most part there is not a lot in it. We have not worked heavily in this area. So to be 100% honest and transparent if it feels faster it is likely either perception or because you profile is cleaner and a lacks things like extensions that sometimes slow things down.

That all said, even though I would not expect a big raw speed increase from Vivaldi over Chrome right now, I do believe that our range of options particularly in the case of tab management could allow for workflows that are faster and more efficient in navigating the internet. In addition as others have noticed we disable (or give the option to disable) the various calls home that Chrome (and even Chromium make), so it is worth considering us for those reasons alone.

P.S. As for the ‘mildly buggy’, whatever that is a reference to, feel free to log it https://vivaldi.com/bugreport/ as this increases the changes we will fix it, benefiting all users.

32-bit Linux support is getting increasingly hard because upstream Chromium/Google do not really care about it. For instance in the latest internal versions we do not have AV1 working for 32-bit Linux. So at some point I am expecting it will become too much effort for us (as we are still a very small team and have lots of other things we are working on) but we will do our best to delay that day as long as possible.

I think that sync is either enabled or you have copied over your home directory or parts of it that include our preferences because we do not do that. :D

It should be automatic now both from SBo or my latest-vivaldi script. :D

[EDIT]: Ah… you saw this. I was too slow!

ruario, thank you so much for taking the time to reply to the posts in this thread in detail. It's fantastic that you've put a few moments aside to go through everything and address some important points. Outstanding work which is a model of what makes this community so special.

I think this is very honest, I was wondering the same thing. I think perception does come into it, but there's no doubting how much faster Vivaldi is than every other browser on my netbook [even Falkon]. Talking of which...

Which is greatly appreciated. When that day comes I think I will just hold onto it for as long as I reasonably can. I am still using Chromium 68 on my desktop as a backup browser, even though AlienBob's latest build is 71.

Is sync automatically enabled on a new install? That would explain it. Nothing was copied over though, unless Vivaldi 'inherited' something from my Chromium install. Either way, I'm not complaining, it made things easier.

We can't automatically enable sync if we don't know your user name and password. :D

You can import things like bookmarks from Chrome but it is not automatic and it does not include extensions or theme options, so something else is going on here.

Correct. So, it must be difficult to know how many people are using Vivaldi?

Oh you mean our own stats? As Lysander666 quoted from our privacy policy before:

Only Vivaldi clients would be making such a request and since each post includes a unique user ID, yes we know how many active users we have. Anyone else however, is likely just guessing. :D

Well, no, as I have read the privacy policy.

I was thinking more in terms of how web sites collect data, for whatever purposes, so that Vivaldi (and others) are properly represented and not lumped together as chrome.

It would appear that, e.g., w3schools,

https://www.w3schools.com/browsers/default.asp

lumps the chrome clones together, but not Opera. OTOH, is that the "new" Opera or the "old" Opera and if it is the "new" Opera, how do they know the difference?

BTW, I read somewhere the code for the Presto engine was leaked in 2017. Has anyone grabbed it and ran with it?

EDIT: not sure if I should have posted that but just in case: yes, it's out there, and not hard to find with a little creative searching [cwizard I know you saw the info].

Worthy of consideration.

I am not sure it matters too much if random site X lumps us in with Chrome

I have seen people quote w3schools before for market share. I am hoping/assuming you realise that it makes zero sense from a point of view of Global market share. Just in case you or anyone else reading does not I shall explain why. Their list is just a list of browsers that view their site. Yes they have a few million hits but in the grand scheme of things that is basically zero (given the size of the internet). Also their sample is highly skewed to the type of people who would look at their site, i.e. people learning web technologies at beginner to lower intermediate level, from countries where the population has a strong command of English. Nobody else visits them.

Consider that traditionally (in Presto days), Opera's strongest market was Russia. At one point Opera was second only to IE in Russia but basically nobody from Russia would visit a site like w3schools, certainly not normal people with no interest in learning HTML. Or look at Vivaldi these days. Our current strongest market is actually Japan. Again, nobody from Japan is going there. So using their site as a means of trying to guesstimate global market share is pointless. Again I appreciate you might not be suggesting this but I want to put it out there in case anyone else makes that (bad) assumption.

The only types of sites with big enough populations of users from a wide variety of regions, who could probably give you some idea of Global market shares are places like Wikipedia, Facebook, and Google and even these sites and not always popular in all large regions. Others who might have some ideas would be large online advertising networks (again Google being an obvious one) or services that are dedicated to stats and have their stuff installed on a wide variety of popular sites. Once more Google (analytics) or companies like Statcounter. Even these are imperfect as they are not truly ubiquitous (and as I said before we lie to Google) but would still be a hell of a lot better than the likes of w3schools. I really cannot fathom why people mention them but they do. Back in the old days at Opera we occasionally talked about market share and people said we were lying because they had looked at the w3schools stats. Which always left me with a ‘WTF’ thought.

Oh and statcounter while far from perfect do have some of there information publicly if anyone is interested, e.g. http://gs.statcounter.com/browser-market-share

P.S. The fact that there are no (IMHO) great public stats is another of the reasons that browser manufacturers keep their own stats. If you know how many users you as a browser maker have and know the current estimated size or the internet population, you can at least roughly calculate you own market share, which is also useful when talking with potential partners. Now of course your potential partners might not believe you but that is its own issue.

P.P.S. Our stats are primarily for our own usage and we do not currently share them publicly (maybe that will change one day). Just to get that out there before you ask me how many users we have. :)

That would basically be impossible IMHO. I am not just talking about the legal issues, though that would stop most groups from trying to do something with it publicly but also the technical challenges.

It is a pretty big code base with a lot of very specialist knowledge required. There aren't many people who know their stuff well enough to maintain this. Sure there are plenty of smart devs in the world that could work it out but it would take a lot of them (coordinated) to do it and time to familiarise themselves. Most of the obvious devs who already understand the kinds of technical challenges required to maintain a rendering engine already work on other browser engines or they are the ex-opera engineers themselves. And most of the ex-opera people moved on to other full time jobs (many for competitors), so would have little time to do this, given it used to be their full time job.

Consider the number of people working on (Core) Presto (when it was shut, not peak) was over 100. Also when the code was leaked it had been a few years with web standards and technologies having moved on. So you would need hundreds of skilled engineers to first get to grips with the codebase and then play catchup. So yeah even with the legal issues aside that is not going to happen.

That said I am sure they are plenty of good ideas and tricks that could have been picked up from that code base that might have found their way into other projects, legally or not.

Well, Opera had something to say about it.

https://blogs.opera.com/security/201...sharing-sites/

It's not that hard to track down more info on it.

Looks like it was incomplete but some people had some fun with it. Some more links of interest:

https://www.reddit.com/r/operabrowse...ble_on_github/
https://linustechtips.com/main/topic...s-been-leaked/

There was something on the Vivaldi forum about it too:

https://forum.vivaldi.net/topic/1343...code-in-github

All --

I am old and forgetful and I often have to look up Saaved Passwords for some sites in my Browser so I can use them 'elsewhere'.

I can't figure out how to view my saved Passwords in the Vivaldi Privacy Settings Screen.

I've googled it ( but not too thoroughly ) and 'they' say Vivaldi was designed to not show Saved Passwords.

Say it ain't so Joe !

Do I have to open Chrome or PaleMoon just to look up a forgotten Password ?

Thanks.

-- kjh( :) please don't say: don't do that because I am also old and set in my ways :) )

Hover one of the website entries and you will see a little eye icon. Click on that and you can show the password for a website that you have saved.

Thanks ruario ...

But I must be doing it wrong ...

All I see is an X Icon which I assume means delete the saved password ...

See the attachment and my Vivaldi Version below.

Thanks again.

-- kjh

Other than a couple of small niggles i have been enjoying Vivaldi as well.

Passwords are one of the niggles, you can access and search them from vivaldi://settings/passwords but i can not find a button in the menu which gets you to this page.

I believe Tools > Settings > Privacy should have a link to this rather than a list of passwords which you cannot view or search.

@kjhambrick Hmm… yeah it seems we only added that recently. I of course run internal builds (the public snapshot builds also supports it). Oh well, a little trick then. We have not completely disabled access to the Chromium settings page and I see that this supports display even in the stable, enter the URL “chrome://settings/passwords” (yes use the URL ‘chrome’ in there rather than ‘vivaldi’, while running Vivaldi) and you should get what you need. But rest assured, it will be fixed and available from the normal Vivaldi settings in the future. :)

Thanks SavoTU and ruario !

SavoTU's link: vivaldi://settings/passwords took me to the default settings page, but the passwords were still hidden.

OTOH...

ruario's link: chrome://settings/passwords took me to the page with the EyeBall Icons -- exactly what I wanted.

I bookmarked it :)

Thanks again !

-- kjh

Thing you might notice is that when you load “chrome://settings/passwords”, the URL presented turns to “vivaldi://settings/passwords” (even though you are within Chromium settings). So SavoTU might have intended to give you the correct link but when copy/pasted it became incorrect.

ruario - could you share a request with your team? Would it be possible to include an option in the settings to change the colour of the window button icon and the one that appears in the native window? I suppose even just a monochrome one would be great, but other colours could be added too. I've currently changed the one that appears in the native window to monochrome, and I switch back and forth between red and monochrome for the window button icon.

Red does rather stand out as a colour which can be a good for finding the browser quickly but it can also be distracting depending on what the rest of one's desktop looks like.

So I have two folders - one for monochrome icons and one for red icons, and I use a command such as

or

to switch them over at will.

I imagine this option could be enabled by just ticking a box in the settings [or menu]?

Its funny how small differences in work flow give different output. If you select then use Go to vivaldi://settings/passwords from the right click menu you are taken to the correct page but copy and paste does not work for some reason.

I wonder if it would be possible for someone to comment on the upcoming changes to Chromium which see it disabling ad-blockers. This is a big minus for all Chromium-based browsers. This leaves those of us using such browsers with two options - hosts files or switching browser to something Firefox-related. It may sound drastic to change browsers over something like this, but ad-blockers are one of the things which make the web usable.

I'm by no means technically qualified to provide an adequate comment, but as an end Vivaldi user (for some reason I had much more luck with Vivaldi's developer tools than with Firefox/Chromium ones), but if this "innovation" of Google makes it into Vivaldi, I'll prefer to abandon it in a blink of an eye and go back to cringe in pain using the clumsy developer tools of Firefox.

I will quote myself from twitter

We are of course discussing this internally but it is early days.

+1 if adbloc is removed from Vivaldi, or if "clean filters" ala microsoft concept are inserted by Google into chromium code, I will stop using chromium based browsers altogether. The INTERNET (post DARPA) was originally designed for all users to exchange ideas, not just the key providers biased agenda. I wonder if anyone remembers that mind thought, with any bias, is the end of freedom. As it is Google has monetized the Internet more than it should be. I started using the internet when it was gopher and host files for the great sites. I can go back to that if necessary. Cheers

All --

I posted a question on the Vivaldi Linux Forums: https://forum.vivaldi.net/topic/3416...of-libreoffice

I posted the Q in the Linux Forum but there were a lot of Windows-Only Solutions and ultimately, the only solution is to change my Global mimetypes file.

Not gonna do it.

So I'll ask here ...

My job is mostly Data Conversions and many times I'll get .csv files or I need to connect to a link to grab a .csv file.

Unfortunately for those of use that don't consider a .csv file to be a Spreadsheet, Microsoft does so that means for all intents and purposes, a .csv file IS an Excel File.

So vivaldi graciously opens the file on my system with Alien Bob's Libreoffice Calc.

And if the content of a column looks like a number, then it is a number and I lose all my leading zeros which are essential -- they wouldn't be there if they weren't important.

Is there a secret option in Vivaldi where I can create a LOCAL mimetypes file ?

Maybe an option to associate file types with an Application ?

And I am not sure I have been given the entire story.

Check this out: http://bear.alienbase.nl/mirrors/peo...ompat32-tools/

Alien Bob's Directory is chock full of shell scripts.

Vivaldi will happily open scripts in the browser that don't have a .sh ext. Example: http://bear.alienbase.nl/mirrors/peo...ols.SlackBuild

But it will NOT open the ones that do have a .sh extent.

When I click [Open] in the Javascript Dialog, Vivaldi tells me that it's too dangerous. Example: http://bear.alienbase.nl/mirrors/peo...ssconvert32.sh

All I want to do is look at the script in the browser -- I would never RUN a remote script which is what Vivaldi seems to interpret as [Open] ...

I can do this in Firefox or PaleMoon -- I set the Application to gvim for .csv files and for .sh files and then I click the 'Always Ask' Box.

Has anyone cracked this nut ?

Maybe a LOCAL mimetypes file like mozilla (used to) have and like PalMoon still does have ?

Thanks.

-- kjh

version 2.3.1440.41 released.

Installed and working here.
:)

Awesome browser. I have been using it for awhile now and i absolutely love it.

CSV (Comma Separated Values) has been an exchange format for spreadsheets and databases since at least the 1980's (I know that dBase, Lotus 1-2-3 as well as SuperCalc supported it).
So it's long been recognized before MS Excel even existed.

Expansion: in .csv files, things with quotes around it are TEXT strings, values without those are numbers (which in spreadsheets often are stored in binary). So
In databases this may be different, you may try to associate .csv files with LibreOffice base instead of calc. Haven't tested that, though.

Well, I just gave Firefox a good old college try and found it to be unstable and wanting in features I use with Google Chrome. Soooo, it is time to give Vivaldi at least equal time for a good shake down. I have to admit that Vivaldi is growing on me. Maybe it will win me over

ehartman --

Yes, exactly. I've been exchanging Data.csv Files since the 80's ( before there was an Excel )

My Problem is exactly the opposite.

I DO NOT want Vivaldi to automatically open Data.csv Files in LibreOffice ( it does do that ).

As you said, not all CSV Files are Spreadsheets.

In my case, NONE of my CSV Files are Spreadsheets.

I want Vivaldi to open them in the browser, or better yet allow me to configure Vvaldi to open the file with gvim ( FireFox and PaleMoon both allow for this ).

Apparently the only way to do this in Vivaldi is to change my GLOBAL MimeTypes File.

Then I would like to look at Script.sh files in Vivaldi.

But when I try to open any .sh file, that means 'EXECUTE' ( inherited from Windows, I am afraid ).

I NEVER want to RUN a .sh file ( or any other Executable ) from within the browser, again inherited from Windows which is why Windows has always been a virus magnet.

Anyhow ... enough with my ranting.

If Vivaldi would fix this design flaw ( no LOCAL MimeType File ), it would be PERFECT for my needs.

-- kjh

Congrats to Vivaldi for coming third equal in the LQ awards this year. Great work!

The percentages look a bit confusing when all listed, so here's a pie chart. One can easily see how Vivaldi's slice is a lot bigger than most.

Thanks. Jeremy posted the pie charts as well (https://www.linuxquestions.org/questions/2018mca.php) :)

Is there any way to get Vivaldi to simply use the system theme? I generally like how it looks, but on the other hand it does kind of ruin the uniformity of everything else.