The context menu does not offer anything useful in my opinion. The possibility to open a pdf-file (to name the most important example) is simply lacking. You have to save and then open the pdf.
Sorry for the misunderstanding, I was thinking about the transfers page, which obviously only work for already downloaded/saved files. From there the files can be run via doubleclick or context menu action. I agree, that the ability to save the file internally to a temporary location and open it directly would be an useful addition. vbr
... - password manager - better download dialog (enabling me to alternatively open the file) - dropdown functionality of the backwards icon ...
Hi, I think, some of the mentioned features are already usable - browsing history in the backward/forward icons is available on clicking and holding left mouse button over them - there are some actions in the context menu (right mouse button) in the download dialog
There is work in progress in some of the mentioned areas (password manager).
I would join the request for an element-wise blocker or a wildcard-based blocker like in classic Opera.
Some minor bug I noticed: when closing a webpage (no matter whether closing it with mouse gesture or clicking the tab's X) and this webpage was displayed in the only/the last open tab this tab stays open but is left unusable. It doesn't react to entries in the address field anymore. To continue you have to open a new tab. The old tab can be closed afterwards. Opera in a similar situation closes the last tab as well which isn't the perfect solution because you have to open a new tab to continue. But it's better than having a non-working tab.
Hi, I remember similar problems in some earlier version, but it seems to work ok for me currently. Anyway, you can set the prefered behaviour in this case in the Configuration manager Preferences: [All settings] In the TabBar section, there is an item called LastTabClosingAction with the possible values: openTab, closeWindow, doNothing. hth, vbr
Hi, I just noticed another feature mentioned previously and wanted to ask about the current status (and perspectives). This is the support for user javascript; there is a notice in the respective github issue: https://github.com/OtterBrowser/otter-browser/issues/336 according to it, the feature isn't that difficult to add per se, the problem beeing an effective url matching mechanism. I'd like to ask, whether it would be feasible, to use fixed settings to domains - maybe via the site settings dialog - scripting - in the same way like user css. A finer differentiation of the URLs could then be delegated to the code of the scriptfile itself. I believe the mechanism of the site settings would be reused, which is already in place, and ideally no additional slowdown would occur (for pages with no userscript set). Possibly, a global userscript could be supported too, if neede - it could be managed via Settings - Advanced - Content (together with user CSS). (I am not sure, whethere one script file for a specific site is sufficient, or whether these should rather be folders, but the logic might be the same.) Would someting like that be doable, or are there other limitations not mentioned in the github discussion? (Please let me know, what is the preferable form and place of such posts - this "Features" thread, a new dedicated thread in this forum, or maybe the github issue?) Thanks and regards, vbr
Which specific Otter version and OS are you using? I'm seeing 100/100 with 0.9.05 beta 5, 32bit (Win 7); however there is a title "You should not see this at all" in the rendering, hence it probably can't be considered a clean success... vbr
... This could be treated more like a fourth level of settings inheritance, right now there are global settings, per site overrides and tab specific settings (these are the most important ones, overriding all other settings) with addition of sharing resources. Kind of like multiple profiles loaded into single window, but with some limitations etc. ...
Thanks for considering this, it really looks like a promising (and unique as far as I know) feature. Of course, tab stacking is useful on its own. vbr
@vbr, what about grouping them in one folder in bookmarks and adding an option to open them as private tabs?...
Thanks, this definitively would be an improvement - however, as I understand it, these wouldn't be recovered on new start of the browser, right? On the other hand, the approach mentioned later might enable this:
@ersi, when grouping will become available then we could consider "private groups", having cookie jar set per group, the same way like is currently done for opening new tabs from private one, they share single, private cookie jar.
Is grouping meant visually - i.e. tab stacking in Opera 12 or would it be somehow internally connected with the bookmark folders mentioned earlier? Certainly it would be an improvement to have multiple controllable "data spaces" for the site data like credentials etc.; a single shared one for all private tabs seems to be "less than ideal" too. For what is it worth, I am now trying a "poor man's approach" with two separate program folders of Otter, both run with --portable flag. The separation as well as the persistence seems to work well, however, it would be of course less handy, if multiple separate workspaces would be needed :-). I bet, there might be some flag for the path of the profile data to be used with --portable, which would certainly simplify this approach; is it maybe the case? are the available flags listed somewhere? Regards, vbr
@vbr, what about disabling cookies for selected websites?
Well, that might work well in general, but it would require determining, which "target sites" are involved, ideally, in advance, before loading them. It seems more straightforward to me to avoid logging into such global service, where tracking the online activity can be expected. But anyway, it can be achieved by using another browser or with private tabs (in the latter case without the persistence). Regards, vbr
Have you tried --privatesession? Is something missing there what you need?
Hi, Thanks for the hint, this flag behaves like the regular private window (which is probably intended). It would need to add some persistence (preferably over tabs rather than windows, but the difference doesn't matter much). Are there maybe some other flags regarding persistence, which could be combined with --privatesession? (I am currently only using --portable, which obviously doesn't matter here.) Those tabs would be recovered after closing and reopening the browser and would keep the site content including state, previously supplied credentials etc. - much like regular tabs do just now (if restoring the tabs is checked as startup option); however this state would not be accessible for sites open in other regular tabs. I could find some references on site isolation/ tab isolation/ process islolation in similar context on the web, but they seem to deal with the stability of the browser with respect to crashes of individual page tabs rather than with the separating of the credentials, I am primarily thinking about. thanks and regards, vbr
Hi, I just wanted to ask for some option in the tab managment, which might be helpful for some circumstances. I think about a permanent private tab; the main usage in my view would be isolating pages with regard to credentials (e.g. Google, FB and others, in order to prevent their linking and interchanging the data with affiliated sites browsed in the same time). I believe, the private tabs (or windows) can manage such isolation already, but it is less handy for regular usage - manually loging repeatedly. I guess with regard to the user interface, it might be sufficient to have a menu option for an open private tab to remain persistent, ie to reopen it with the previous content after a new start of the browser (assuming this is set as the startup option generally). Of course, there are more complex possibilities too (e.g. a dedicated category of an "isolated tab" or window respectively, which would behave like this, these could be saved in bookmarks with this additional setting etc. As far as I know, inorder to achieve this, one has to use a separate browser currently, or am I missing some existing options? Do you think, it would be doable and interesting for the users? regards, vbr
Hi, for what is it worth, I just checked the sample page with rotating title text via javascript. On windows 7 I don't see Otter crash or become sluggish while running this script (within about an hour I tested it), However there is some regular increase in disk activity. I is audible, but not noticebly invasive in any way. The procesor usage of Otter is between 0-2,5 % with this rotating title; for comparison - scrolling a simple static web page up and down shows about 5 %, for the Gmail inbox such scrolling needs some 15 % processor usage, regards, vbr
