I don't see an option to change the view. Does this mean I need to be signed in to do that? If so, that's a really poor design. I should be able to delete photos just as easily whether or not I'm signed in. I'm not sure why the toolbar UI needs to change depending on signed in status.
The Trash option available while signed in vs. Deleting locally mean different things. The first is moving the file to trash where it can be recovered for a little while, the second is blowing the file away from your file system. We wanted to keep these features distinct and clear.
Though I can understand for those using the app entirely signed out this can be frustrating. Thanks for the feedback. I do hope you sign in however, there are a lot of things available to you signed in that simply don't exist in the signed out view.
It would be useful if the app made the differences clearer - you can see the confusion in this thread alone where people just can't see why my delete icon is missing and theirs is present (and the ensuing downvotes because it looks like I'm just doing something wrong).
I have always used the app offline, so it's quite frustrating that a major part of my workflow has been made more difficult. I'm not sure what the benefit was in changing this behaviour for offline users - there's always been a trash/delete icon. Now someone has just taken it away.
Ideally, I'd love to see it return and the confirmation dialog to disappear (or be configurable with a preference). Or if you're using it offline, add a confirmation dialog ("You're PERMANENTLY deleting this") with a "Don't show this warning again" box. And possibly add a "Deleted (Undo)" toast (delay actual deletion on the filesystem by 3-5 seconds).
I think most users understand the semantics of deleting a file, and if this change only affects offline users then the affected users already know that delete means delete forever.
I would actually consider signing in at some point, but I now feel my hand is being forced just to regain some fundamental functionality.
Additionally - what about users who can't use the app online? For example users with limited data packages or users who travel a lot (where roaming data can cost upwards of $5/MB).
Chrome has a concept of multiple users in it's preferences. When you turn it on then you can pick an icon per account and it shows up on the top right of the browser window. To switch to a different user click on that and switch to a different user. Each user can sign into a different google account.
Each user has it's own cookies, chrome extensions etc. Keeps work and personal accounts very separate.
I started using this a few months ago. It definitely helps a lot, but it can still break the workflow.
Using OSX, when I click on a link inside Mail, it opens it in Chrome using the last used user. It's annoying when it's the wrong user. You still need to switch users and reopen the link in Chrome.
This is why I use different browsers for different accounts (Chrome, Safari and Firefox are generally all fine for regular web use) and Choosy (http://www.choosyosx.com/) to prompt me which browser to open the link in.
It also allows for binding certain URLs to specific browsers.
I have the same problem, as I have a personal and a work Google account. Chrome can sometimes ask you if you want to switch accounts, but in the end, I use Firefox for personal stuff and Chrome for my work.
To sync to Google Drive, I use Syncdocs http://syncdocs.com which is an enhanced Google Drive sync app that enables syncing multiple accounts at the same time.
I think Google wants each user to only have one account, which makes it better for them tracking ads.
I'd recently looked into ways to cure the pain of multiple google drive accounts running on the desktop app. People spoke highly of CloudFuze and InSync, but their support forums were filled with really awful error reports about random file deletions or endless file replication.
These services sound like a syncing layer on top of a syncing layer, which seems prone to errors.
It's a fucking godsend, every computer I have has 3 different Chrome accounts on it, all short cuts on my tool bar.
On windows at least it is trivial, extremely trivial, to switch between them. You will generally be in one of the modes, for me these are "their company", "my company", "personal".
Whichever one you used last will be the one that receives the "open internet" command. So if you are in "their work" mode, reading your emails in their email client it will open the google doc in the right context.
For you your workflow is somehow broken, for me that's exactly what I want. If I'm in "their work" mode, I want all browsers to open in that context. I'm not expecting them to be psychic. That means 95% of the time it just works because you are already in the right context. The other 5% you get used to very quickly. If I open up my personal email account, I want any next links clicked to be opened in that context.
Added bonus, it's great for every SASS not just just google accounts.
Would be cool if it could detect which users had access to the document and switch to the most recently used of those to open the link. Not sure if that can be done securely though.
Firefox has an even more powerful version of this.
If you start firefox with the "-P" flag you can choose to create a new profile. You can also pass it an argument (e.g. firefox -P default) to choose one.
In this case, the profiles are completely disparate; there is zero overlap. In this case, you simply have to login to one google account per window and paste into the correct window (still not ideal).
To run multiple profiles at once, launch all profiles after the first one with "firefox --no-remote -P <profile-name>". Clicking links will open them with the firefox that was launched without "--no-remote".
How is it more powerful? This is exactly the same thing, but apparently with a worse UI (the multiple profile thing in Chrome is exposed through the UI).
The firefox instances have completely different processes, settings, etc. Everything. Chrome, unless I'm mistaken, does not go that far.
For example, one of my main uses of multiple-profiles is that I have a different profile for every proxy I use. I can launch a firefox profile that's proxied side-by-side with my usual firefox (aside, the firefox proxy settings are exposed via the UI, unlike chrome).
Chrome, I'd have to run "google-chrome-stable --proxy-server=$proxy" and then, again unless I'm mistaken, all accounts will use that proxy server. That, by itself, is a deal breaker.
I'm not familiar enough with chrome's settings and so on to say what does and doesn't leak; I could be completely wrong on all of this, but I suspect I'm correct.
Edit: On looking more, Chrome's does look more complete than I thought. The proxy bit still is a dealbreaker for me (well, and I'm adverse to logging into a google account), but I retract much of what I said.
I haven't used it in a long time, but I did use it at one point.
It's just not as good security and privacy-wise. Using proxies on/off on one profile, as it encourages, results in any tracking cookies seeing both IPs having the same tracking data, and thus your proxy has lost some of its privacy.
I also run entirely different extensions when I'm going for privacy vs fun browsing vs banking etc etc.
If you just want proxies to get around some region restriction and don't really care about the privacy or security aspects, then FoxyProxy might be fine.
I wrote a Firefox extension on 2007 that allowed you to use different accounts at the same time in different tabs (no windows!). The video showing it continues to be available: https://www.youtube.com/watch?v=1EBkB-Yp-zM
> "Keeps work and personal accounts very separate."
Does it? I don't know either way, but if I were google I would link the two accounts (just takes one more cookie at most!) so that if me@work-acct searches for a car, then me@home-acct can be served a car ad....
Yeah, chrome profiles is the answer.
I typically have two chrome windows open, one with my work account and one for personal. Each is linked to the corresponding Google account. This gives the advantage of chrome sync (bookmark sync, etc.)
That said, the OP had 4 google accounts, so I can see how this could get messy.
The last (and only) time I used this it ended up adding all my development-focused extensions to my SO's Chrome profile, then when she removed them made a mess of my Chrome configuration on other systems, so YMMV.
Here's some more info about Chrome profiles for whoever is interested (http://www.googlegooru.com/how-to-create-google-chrome-profi...). Like most others have said, I have three Chrome profiles on every computer that I use. It's perfect for maintaining your workflow across multiple devices, as you can open up the same tabs at home that you had open at work.
People here have given great motivating responses. But In don't think anyone has answered the whether it would be hard to get an job again after a big gap if you decide to get one again. It seems like you want to continue coding in some form, so really there will be no gap of experience in your resume.
If anything finding people who have built products themselves and put them out there is great to see and are hugely valuable to an enlightened organization. Whenever I interview people this is always a fun part of the conversation. I find that those people are very likely to be tuned to building great products and are the kind of people I want to work with.
I can easily imagine what some of the dependency management issues could be, but I'd love to hear your thoughts on the following two options:
In the Java world there is a concept of a .war file which packages everything up so it's easy to deploy on app servers that handle it like: Tomcat or Jetty.
Also what's wrong with giving out a prebuilt virtual image for a single server install? Github enterprise does this.
I'm not a fan of app servers; I'd rather segment my running applications. I use maven-shade or sbt-assembly to create an executable .jar with an embedded web server. (Also makes it easier to just provide a runtime interface with jsvc.)
Also makes it a hell of a lot harder to scale horizontally. With an app layer you can just horizontally scale your app server stack and put a load balancer in front of it and you're done.
I don't know where you're getting your information, but your claims do not match my experience or, apparently, that of pretty much anyone else using Play (embedded Netty) or Dropwizard (embedded Jetty with a handler). Or anyone who is writing a Rails app or a Django app or anything else that lives as a standalone HTTP handler.
I have a Puppet script that sets up deployment on a new server. Puppet does not care how many machines the application is running on. Neither does the load balancer/reverse proxy in front of it; it takes a list of IPs and does its thing across them.
I have never seen an app server do anything that wasn't essentially this process, only with more XML and aggravation.
This whole situation makes me think: is there a static analysis tool for excel? Seems like missing a few rows in an excel calculation could be flagable.
On Android push notifications are done through GCM (Google Cloud Messaging), when it's sent your application gets a callback (Intent). So it is up to the application to decide what to do with it, whether it is to show a notification or do something else entirely. So on Android you have a pretty good idea that the the user is actually getting a notification. Jelly Bean and up you can also do very rich notifications with images and actions built right into it.
On iOS when a remote push notification is received there is no indication sent to the application at all. The only way the application knows something has happened is if the user actually opens the notification.
I haven't programmed on windows phone, so I am not sure how much an app developer knows that a user is viewing a live tile or not. Maybe someone who's built one can chime in.
Overall it seems to me, that what the author is describing is Android, where the app developer has at least some feedback as to whether a user is seeing the notification (albeit not an exact: user has seen x notif).
Only if you use their inbox-like thing, which will still not tell you deliverable rate, it just ensures the recipient can see a list of all attempts.
The actual push notification itself is entirely untrackable unless the recipient actually taps on it, or it is received while your app is running. IIRC you can read out the application's badge number, but that only shows you what the most-recently-received notification was (if it set a badge value), it implies nothing about any intermediate ones. There is no way to list the notifications, nor count them. You can do this with local ones you generated yourself, but not ones pushed to the device.
While someone is actively in your app, you do get a method called that gives you the notification as it arrives - which is handy, because the system entirely mutes it (doesn't go into the notification center, doesn't appear at the top of the screen). You can do that to do a negative tracking system ("they didn't receive this, don't count it in the stats"), or to show it yourself.
The usual approach is to track acks or callbacks driven via user actions on the server-side. Adding a "guaranteed delivery" QoS will involve queuing, waiting and retrying on non-ack timers - and do so without abusing users' trust in your app.
