Maybe I’ve kept the Cr-48 a bit too quiet in the past week, but barely 90 minutes after my last post, I restarted Chrome OS into a new version. The little gripe about the timeout when checking Chrome OS version has been resolved, and there’s a fresh new Google Chrome build in place, too.
On the Cr-48
Chrome OS build 20.0.1116.0 dev
Chrome OS Platform 2199.0.0 (Official Build) dev-channel x86-mario
Cr-48 Firmware Mario.03.60.1120.0038G5.0018d
WebKit 536.9 (@115072)
It’s been out and about for a week, but only yesterday did I dive in to Chrome on Android 4.0.3 / Ice Cream Sandwich / ICS.
Monday’s lecture class went swimmingly, and I lectured for three hours with the Cr-48 displaying three different slide sets. Everything back to normal, performance-wise, with multi-page PDF display. Cr-48 battery performance continues to be excellent.
Today at work I reached a threshold of need where I truly just wanted to compose a multi-page PDF out of a pile of single-sheet map graphics, and I finally made use of PDF Split and Merge from pdfsam.org — after downloading from Sourceforge and finding it clean with AVG virus scan. It was awesome, and motivated me to create a cover page in a presentation program (OK, it was MS PowerPoint 2010) and then printed the single page in B-size (11×17 inch) landscape by using Cute PDF Writer to control the layout size. Thanks to PDF Split and Merge, I could easily work out an arrangement where two-sided color 11×17 landscape printing made a very fine booklet to allow side-by-side comparison of related analyses (present and future viewsheds). The prints were pretty good when stapled into book form, but the multi-page PDF display was extremely effective thanks to the precise scale control afforded by geographic information systems (GIS) software. With the multi-page PDF, a simple PageUp / PageDown flick of keys flashes the two analyses for current and future visibility, and allows one to study the changes in considerable detail.
Enough with the PDF viewing thing.
The latest efforts now are to keep up with the Chrome updates. At work and at the college office, I now use Windows 7 machines; one is a mighty nice new Xeon W3680 with 12 threads and 12 GB of memory, the other is a nine-year-old Pentium 4 HT 2.8 with 2 threads and 4 GB of memory. One of the home machines is a Windows XP (Media Center) SP 3 with a seven-year-old Pentium D 820 with 2 threads and 3 GB of memory. Throughout each week I either manually update or let background updates happen to Chrome Canary build.
A bit bigger home machine is running Ubuntu 11.10 x86_64 on a five-year-old Core 2 Duo E6550 with 4 GB memory, overclocked at 3.3 GHz now, although for its first four years it ran fine at 4.2 GHz with air cooling. It’s wired up to run the latest unstable Google Chrome build that gets pushed out on dev channel for x86_64 Linux.
And now this week, my ICS phone has a beta channel Google Chrome running on it. It’s pretty sweet getting a straight-up tabbed interface that is fairly consistent among the various Google services. I really like the simple way that Chrome on Android is handling the dismissal of deprecated tabs. Overall it might be slower for some page loads than Android Browser, but its consistency of interface makes me more than happy to use it in preference. Today I dumped Android Browser and Messaging from my action bar (whatever the best name is for the five icons along the bottom of the Android screen) and replaced them with Chrome Beta and Google Talk.
Also in the mix are a couple of Sony Intenet TVs runing Android 3.2 (Honeycomb)
And as typically, my Cr-48 got an update today. So here’s the whole spread:
Windows (7 and XP)
Google Chrome Canary 19.0.1042.0
Ubuntu (11.10 x86_64)
Google Chrome 19.0.1041.0
Android (3.2 / Honeycomb) Google TVs
Google Chrome 11.0
Android (4.0.3 / ICS) Nexus S
Google Chrome 16.0.912.75
App Version 0.16.4130.199
WebKit 535.7 (trunk@104610-dirty)
Chrome OS (Cr-48)
Google Chrome 18.0.1025.32
Platform 1660.34.0 dev-channel x86-mario
WebKit 535.19 (@107639)
So there’s ever more company in my Chrome world. The Canary builds might not sound too interesting, but they pull my interest forward and often provide helpful improvements to my usage. Another aspect here is that I’m running four different user profiles, as a means to separate work from teaching from personal / hobby usage—not to separate platforms! So I’m making almost daily usage of three Google accounts on all platforms. In fact, I use yet other profiles to set up Google TV, but those are fairly static. Since my phone got up to Android 4.0.3, I’ve been really pleased with the way that multiple Google accounts are handled by the Gmail program, and I’m getting very used to the way that Chrome 19 will spawn a new browser to handle different accounts open at the same time. Google Docs still drives me to log out of the less active profiles to get write access.
After my phone upgrade to Android 4.0.3, but before Chrome arrived there, I was already appreciating the way that my photos and (with WiFi) videos just appeared up in Google+ and that took care of a lot of sharing. All the same, in the past 10 days I’ve started to make much more use of Dropbox.com for sharing work product. That experience is pretty good across Android and the places where I use Chrome browser.
About a year ago, I filled out some fortuitous web page, was one of the first two in my Zip code to do so, and bang! The next afternoon, there was this awesome little package on my doorstep. This morning, I found myself contemplating cosmic alignments both large and small. First the large:
Then the small: with great good fortune, the Cr-48′s arrival coincided with my returning to some part-time college teaching responsibilities, on top of a regular full-time job with our local county making maps. As a geophysics graduate student, I started building customized hardware about 25 years ago; the insights gained then through some strenuous effort have rooted into a deeper-than-average understanding of desktop, workstation, and small server technologies—specifically when related to physical modeling of geospatial phenomena, digital cartography and imaging, and applications of photogrammetry or the measurement of quantitative things from photo images.
Thanks in part to my mix of paying work during 2011, I’ve found reason to use Chrome OS, Ubuntu, Android and Windows nearly every single day. My workplaces do not provide them, and I choose not to afford Apple products beyond an iPod Touch; I spent my time in the Reality Distortion Field between 1990–1994 using, promoting, and being trained to develop for NeXT workstations—and it’s those innovations that I perceive add most of the value to the OS side of OS X and iOS. I saw how much Apple focused on getting their products into the hands of students, and supporting the app environment; I’m not writing about the iPad that my kid’s using in Kindergarten—I’m writing about the Apple II and the lock it had on educational software.
I was there, watching a Super Bowl game with friends and fellow students at Stanford, when Apple aired the flagrantly copyright-violating “1984″ TV advertisement announcing the Macintosh. It was an innovation, and its interface was attractive even while its performance echoed my Osborne 1 dual-floppy system. But that was a different time, when professors still had secretaries to do their typing, and the Macintosh, when it was embraced by someone besides a grateful administrative assistant, was the first choice of those who were uncomfortable with computing devices of the time. Apple built on this market to bolster the confidence of its Macintosh users with advertising images that conveyed how hey, even if you’re still a bit uncomfortable with computing, you’re really cool because you’re using a mouse! From those years to the present day, my oft-validated perspective is that mainstream users of Apple computing devices (professional graphics, music, and video editors aside) are spending their way to consumer coolness rather than building their skills to output creative greatness. I can understand why people want to have the computer get out of the way of their communication with friends and consumption of media; for most of my work, it’s important to have strong facility with the device, and be comfortable with several platforms. ’nuff said.
I’ve spent quite a bit of time scratching my head about what Chrome OS is for, what it’s all about, and why the Cr-48 should exist.
For about a year before the Cr-48 showed up, I’d been using an Android reference platform, the T-Mobile G1, migrated up through Froyo using Cyanogen Mod. About eight months ago, I grew tired of stability challenges on the modded G1 and bumped up to another reference platform, the Nexus S on T-Mobile, now rooted but not ahead of T-Mo release versions of Gingerbread. My use of Android phones, and appreciation for how the feature set has grown from excitements about Cupcake updates to now were a source of interest in the Google TV platform. So I got first one, then a second of these Android Eclair devices with big screens. They are both updated to Honeycomb at this time, and we’ve kept the cable subscription at the sub-basic level with only local broadcast and public access channels. Everything else, content-wise, is from Netflix, Sony Qriocity, Amazon Video, or a few DVD and BD discs. I don’t sense a lot of overlap between the TV and phone device uses of Android, but I do recognize that related versions of the same OS are doing moderately different things pretty well in both cases.
But it’s my college teaching job that has really catalyzed my understanding of how Chrome OS makes sense. There, I sometimes need to create maps for lecture slides. In a few local cases, there are specific county data sets that I’ve already worked up using Windows 7, some maps I make at home using ArcGIS and Windows XP, some Google Docs presentations prepared on my home Ubuntu workstation, and last-minute changes from the Student Union using the Cr-48. Depending on the lab I’m teaching in I’ll either use the Cr-48 itself hooked to a projector, or install Chrome SXS / Canary on the instructor Windows 7 workstation that is already hooked to the projector and either run Google Docs from there, or download a PDF and present from that if there’s any problem with connectivity.
So thanks to my teaching work, and having to fit it into spare moments around a regular full-time job, I have really come to appreciate how the cloud makes me more efficient at preparing lectures. Whatever workstation or machine I’m at, I just use it for whatever it does best, and the results quickly build toward my final lecture. I believe that it takes me only 1/3 the time to prepare a lecture than it did when I was exporting presentations in various MS Office formats, and carrying them around on a thumb drive, or posting them for download, or e-mailing them to myself as attachments. Now, using Google Docs, I am effectively collaborating with myself when I’m working from various different contexts.
And although it’s not concise, that context is why Chrome OS makes sense to me. Sure, when I think of the folks at the Googleplex and around the world working to move Chromium and Chromium OS projects forward and product-ize it into browser and OS forms, there is a warm spot in my heart, for the insight and rationality I sense in the directions chosen that makes me think back to Stanford where some sense of tribal affinity is felt, a sense that I do not hold in a comparable way for teams centered at Cupertino or Redmond. I’m reading or at least searching news items related to Android and Chrome OS daily, and I try to maintain a current sense of where at least some voices think these projects are going.
But my take seems just a bit different. Perhaps it’s the affinity that I mention above, where I’m really expecting some underlying insightful if not brilliant architecture to emerge. That’s the template that I fit news items against. Sometimes things related to Chrome OS really aren’t all that great.
Like why, after all the agony caused by the oversized track pads, haven’t Samsung and Acer developed trapezoidal ones, rather than rectangular ones? Seriously, even if the touch sensors aren’t denser at the narrower top end, the response could be adapted in software by an affine mapping of the narrower-at-the-top trapezoid into a rectangle for feeding the OS touch events…
But I’m immersed in at least a weekly cycle of Windows XP, Windows 7, Ubuntu, Chrome OS, and Android. In all but the phone environment, the Chrome browser is the common thread. And that’s where it all makes sense to me. My insight was piqued within the last month when one day (and it was really only there one day, as best I can tell) the Chrome Canary browser displayed a little green up-arrow icon by its wrench telling me that it might be time to update. Whoa. That little icon broke down a mental barrier that I’d been holding up between my use of Chrome OS and the various contexts of Chrome browser use.
Then it all came in a cascade of “oh yea…” experiences. Like the day that I had a technical support challenge at work and wasn’t configured for VNC access yet one of my colleagues, knowing me as a Chrome chauvinist, suggested that we solve the problem using Chrome Remote Desktop. I installed the extension and it worked perfectly. I’ve since used it behind and through firewalls. And Google Cloud Print has been important enough to my use of the Cr-48 that it has attenuated my use of Chromium browser. I’m willing to be 72 hours behind Chromium and running Chrome SXS just so that I can run Cloud Print. In a way reflecting all of the cloudy goodness that makes it possible for me to fuse graphics from many different platforms into lecture slides, having any old machine that I’m logged into and running Chrome able to serve up its printer connections is wonderful in an over-the-top sort of way. I mean really, I’ve never ever had a seamless cross-platform, cross-site unified list of output devices before, and yet there it is, all available to me with the Cr-48.
In the past year, I’ve never had the Cr-48 plugged into the home network via cable—because I can’t! But there it is, whether it’s using a personal router at the college, some sort of wireless connection at work, or is just in the vicinity of the home network, it’s as if I had Unix CUPS running on the Chromebook. But I don’t need to. As long as I’ve worked out the printer settings for whatever machine I’ve got going at different workplaces, they are ready for me to use from the Cr-48 whenever I’m close to the printer.
The last piece of the puzzle that helps me feel that Chrome OS makes supreme sense as a reference platform is multiple Google Account profiles. Again, this was prompted by my teaching work. Of course the college has a Learning Management System (LMS) that offers e-mail, class web page, and the ability to post files. But I wanted to help the class learn more about using the cloud, so I have strongly encouraged them to create their term essays not just in electronic form, but in the cloud, shared with me rather than sent as an e-mail attachment.
This meant that I encouraged some students to try out Google Docs who had not used it before; they were not required to use Docs for its document-composing applications if they did not want to, as long as they uploaded their report to their Google Docs account and shared it with me before the deadline. Those using MS Word 2010 were sometimes uploading their reports in .docx format, but over the past couple of weeks, those are now just opening inline within Google Docs, almost no matter what their size. In this way, I was able to accommodate students using lab workstations, home Windows machines, home Macs, or even home Linux. But to manage the deluge of reports, I created a separate Google Account for myself, specifically to interact with the students and their reports.
(There’s much more to be said about what that means in terms of Google Voice for each account, and the way that Android provides a control panel for the multiple accounts’ Gmail and Google+ streams, but that’s for another day)
And that brought me, in the last month, to a much clearer awareness of the nature of multiple login profiles, the partitioning of services among different Google Accounts, and ways to architect free cloud services into a system that supports my needs as an instructor, as a county worker, and even as a household member. With Chrome OS, I can get some functionality across accounts for viewing, but write permissions are most reliable when only one Google Account is logged in on all tabs. With Chrome SXS browser, each different Google account opens in its own browser window with its own set of tabs, and to a greater extent each window is its own sandbox across the workstation’s dual screens.
To summarize what feels like insight, a common-sense perspective from my specific use case, Chrome OS makes infinite sense as a reference platform for the OS that is growing inside the Chrome browser. Remote Desktop, Cloud Printing, and multiple simultaneous logins mean that the basic Chrome OS is evolving within the latest browsers. In case “latest” browser doesn’t make specific sense, for the last three weeks or so I’ve been running updates in the range of Chrome 17.x.x.
To me, it seems rather profound to have OS-style functionality growing inside the Chrome browser. After all, aren’t there perfectly good versions of the Chromium browser in OS X, MS Windows, variations of Linux, on x86, x86_64, and ARM processors? The memory footprint of Chrome is not so heavy that it screams “sandboxed OS!” to most users. But as web sites grow the functionality of their apps toward desktop-grade product, as Javasript runs faster each quarter, then what happens?
I’m suggesting that what happens is that most of the functionality that I’ve been living with as the Cr-48 has evolved over the past year will be existing, implanted in the most popular desktop systems worldwide if the desktop has Chrome browser installed. That means the Cr-48 is not so much a reference platform for Chromebooks as it is the reference platform for the OS-in-a-browser that will just be there, all over the place, ready to consume upgraded web apps. Right now I experience it with WebGL in Google Maps, but what of a browser-embedded OS able to run Native Client apps? Then as functionality of cloud web apps grows, performance may grow right along with them in a very appealing way. This week I saw the amazing work at architectstudio3d.org being used by third-graders—and it gave me great hope for children learning spatial literacy, but it required the Unity game engine plugin; if NaCl in Chrome makes that happen, I will be a happy camper (and parent).
Downsides could be that the Chrome browser runs away with fancy functionality that won’t track with WebKit or V8 alone. I really enjoyed reading the post last week by Nexx positing Chrome browser as the next IE6 and it made sense. IE6 was disruptive in its time because it was fast and capable, but became a problem when it diverged from standards supported by other browsers so that web sites contorted themselves to work well in IE6 at the expense of working as well for most other browsers. I’m not certain that all angles of Nexx’s analogy are equally sharp (in the sense of being pointedly applicable to Chrome), and yet I’m not seeing any bright line separating Chromebooks in total as a reference platform for Chrome the OS-in-a-browser.
Product-wise, I’m still of a mind, same as six months ago, that a transformative and necessary price point will be to have at least a two-threaded, 16 GB, 3G wireless Chromebook available at $249 including the 100MB/mo “free” 3G data. That is a space where wallets will open, popularity will swell, and Chromebooks will find their place in the world of devices. It’s also a space that could be invigorated by student communities who want to produce cool stuff more than flaunt a cool and pricey device—and who are knowledgeable enough to gather their thought-nutrition from the Internet directly. In that world, at least from where I write, iTunes would be on its way toward looking like Prodigy of the early 1990′s, yet another attractive walled garden where the price of admission is an open tab on your credit card.
This post, like most of the ones before it over the past year, have been pounded out on a Mario keyboard. As of today’s update, its tagline reads:
Mozilla/5.0 (X11; CrOS i686 1412.7.0) AppleWebKit/535.11 (KHTML, like Gecko) Chrome/17.0.963.2 Safari/535.11
Or as I’ve usually posted it:
Chrome OS 1412.7.0 (Official Build) dev-channel x86-mario
Google Chrome 17.0.963.2
WebKit 535.11 (@102287)
One year on, and grearing up for the next, Happy First Drop Day, Mario (the day that FedEx dropped you on my doorstep)
Yesterday I noticed that an update had arrived, but was very busy at class and waited to restart the Cr-48 until this morning.
Had a very interesting multi-color fade-to-black on the way out towards restarting. I’ve never noticed that before.
Probably a good thing that I waited, because the login screen changed, and apparently so did the system’s desire to have a live Internet connection at the moment of login, even when I’m logging in to the same account.
In my use since December, it has almost always seemed to be the case that the Chromebook must be keeping a hashed/encrypted local copy of my password, because I’ve had the distinct impression that it could validate a local password entry before the Internet connection was live. I figured that it performed forward encryption of what I typed, and compared that against the local forward-encrypted copy of the password in traditional Unix style.
The new login screen does a rather reasonable job of mimicking a Windows 7-or OS-X type screen, presenting a couple of icons for the local users who have used it in the past. For the last few months, I’ve only noticed my own login, and expected that switching users would require a click. Not at this build.
What I’ve noticed different is that 1) the keypad was not sensitive to tapping by touch—I actually had to hard-click it to signal my choices; 2) The Chromebook demanded a live internet connection at the time I was clicking my icon to log in—and it told me how to do that by clicking the connection icon in the upper-right part of the screen; 3) I actually had to make a valid choice and wait for the connection before I could proceed with my login.
This isn’t so bad if it means that the local hashed copy of the password is gone—I guess. But what about those use cases where I want to use the Cr-48 offline?!? I’m getting rather used to saving important presentations in PDF form and then hitting Ctrl-M to bring up the file browser and launch it from there, without need to hook up the wireless router that I now carry around in my Cr-48′s travel bag. (That file browser got much, much better now that I can delete items from within it.)
So, I’m a bit intrigued, but somewhat concerned that having Internet as a hard requirement for logging in my usual account could get me wedged into a bad place. It will take a bit more checking out to see how it goes.
Now running Mario.03.60.1120.0038G5.0018d firmware, Chrome OS 1011.11, Google Chrome 15.0.874.12, and V8 220.127.116.11
Another week, and new dev builds dropping into the Cr-48. This morning I caught up with the weekend’s states and noticed that for the first time that I’ve noticed since last December, the Chrome OS splash screen has gone white-background instead of glowing logo on a black background.
Also, we’ve broken build serial 1k and now have dev channel Chrome OS at 1011.0 with Google Chrome 15.0.874.5 from WebKit 535.2 of build @94747, with V8 18.104.22.168 – the firmware is at Mario.03.60.1120.0038G5.0018d
Now I must get back to class prep!
Now sitting on the restart from the third update. Chrome OS 587.40 Perhaps this is what would have been called “0.13.587.40″ using the naming conventions of earlier this month. Maybe by dropping the “0.13.” I should take it that the post-2011 06 15 epoch means that we’re no longer in beta?
Google Chrome 13.0.782.34
WebKit 535.1 and V8 22.214.171.124
Time to take the Cr-48 on a little trip. The plane probably won’t have WiFi.
I’m pleased to reflect on the past six months with the little Cr-48. It really has fit in nicely to my computing world. Sometimes I use it at work to stream audio to some noise-cancelling headphones, to focus on some work while technically remaining in compliance with a “no streaming audio” policy. I think that I’ve actually bought a Verizon GB two times in the past six months. When I do, I am well and fully in compliance with the policy. I remember one month, close to the end of my 30 days, when I tried so hard to listen to audio and use up the prepaid data that it became a distraction.
In the months that I have had the Cr-48, I got a pair of Nexus S phones for the household. Those are fun to have—and not just for the streaming video calls that they can make to each other. It’s been refreshing to upgrade my G1 running Froyo to the Nexus S with Gingerbread; very little interface disruption, but a lot spiffier responsiveness.
But for all I read, and read almost every day on the news search terms “Chrome OS’ and “Android”, there seems to be a couple of portals of awareness that people pass through on their way toward a bit clearer understanding of this Chrome OS world.
Has it been one or two updates this week? Anyhow it’s up to 0.13.587.5 with Google Chrome 13.0.782.14 (still no Chrome 14 browser for the Cr-48) WebKit 535.1 and V8 126.96.36.199
Although most of the responsibility appears to lie in the household router, things looked mostly good from Ubuntu 11.04
And things almost looked the same on Chrome OS
After a weekend on the road to SoCal, 0.13.558.3 Chrome OS with Google Chrome 13.0.755.2
For the first time that I recall, the system actually prompted me for a wireless network password before I logged in.
But wait! There’s more! Another update installing. I’m intrigued by the news of another Easter Egg somewhere, somehow connected with the Cr-48 moniker. Shame, but it doesn’t involve pushing the ‘c’,’r’,’4′,and ’8′ keys simultaneously… ;^)
And…poof. There’s a new version in the house. 0.13.587.2 Chrome OS with Google Chrome 13.0.782.3
WebKit 535.1 and V8 188.8.131.52 to round it all out.
I like the nascent ribbon along the bottom of the new page view! It’ s always like I doubt myself: was this here last week and I just didn’t notice it? But the switching between Apps and Most Visited is attractive.
Digging around for eggs does yield some graphics; a little VT-2 and ‘find’ action is all it takes.
Best guess on the Easter egg thus far: hitting Crtl-Alt-Shift-W will make some springtime tint colors appear on the screen.
Pointed out by Richard Preece in the Chrome Book Pilot thread