tecosystems

Miscellaneous Pre-Travel Tidbits

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

Haven’t gotten to a couple of the posts I had planned this morning, first because I got up late after being up till the wee hours of the morning playing with Zmanda, and second because I was out of the office for a client project midday. As I think I’ve mentioned before, I’m out of the office tomorrow and Wednesday for VMWare’s VMWorld conference out in LA. I’m hopping a late night Frontier flight out there, and have quite a bit I need to accomplish before then so I’ll have to wait on some of the planned entries for the day. But a couple of things possibly worth mentioning:

  • Adobe Presentation Offensive?:
    A couple of you have asked privately about my characterization of Adobe’s presentation at last week’s Zend Conference as “borderline offensive,” and I suppose a remark like that does require some explanation. So in a nutshell, I found a variety of the messages very problematic. Not only am I tired of hearing from Adobe and other providers about “how the web is broken,” when it is self-evident to me that it is not, I also didn’t appreciate tidbits like “Javascript sucks” or “the back button not working is a myth.” The presenter also wasn’t winning me over by explaining how Flash is not only highly penetrated but far more frequently updated than browser platforms – not after I had to wait two years for an up-to-date Flash Beta on Linux. The audience didn’t seem overtly hostile, however, so it could be that my reaction was atypical.

  • Ads?:
    Some of you have inquired as to our plans – if any – for advertising here on the site. Brandon‘s comment is a good example:

    I have always thought the Redmonk blogs collectively were very similar to Techcrunch, GigaOm, and others. Only the focus is on OSS, Enterprise software, etc. Have you guys considered at least partially adopting similar business model?

    The answer is yes, we have, and will continue to. We’re not in the ballpark, or hell, league, as Om or TC traffic-wise – nor, frankly, do I hope to be – but advertising is a supplemental revenue source that we are indeed actively considering. Some of you might have noticed that James is sporting a nifty little graphical ad on his blog, while I’ve cut over from AdSense on my blog’s home page to Text Link Ads. We’re also talking to folks like Federated Media to see if we fit their profile. Are we going to get rich and retire off ads? Unlikely, at least any time soon. But neither are we anti-non-traditional revenue strategies.

    For those of you that might be getting concerned, rest assured: everything that’s an ad will be marked as such, and we’ll keep them as non-intrusive as possible. Suggestions or concerns are, as always, welcomed.

  • Speaking of Zmanda:
    As planned, I gave Zmanda’s ZRM – the MySQL backup utility – a whirl over the weekend. Why? This HowTo lowered the barrier to entry effectively to zero. Or would have, if I was on Red Hat or SuSE, as Zmanda only distributes their product in RPM or source formats – not .deb’s for those of us on Debian or its derivatives like Ubuntu. Fortunately, Falko Timme had run into the same problem and wrote up a great how to for converting the RPM into a more apt-friendly .deb file using Alien. Still not ideal, but a better workaround than compiling it from scratch (was surprised to see that ZRM wasn’t in Gentoo, either). From there, it was a snap to install the package using dpkg and get going on the original howto, and true to their promise I was backing up MySQL DB’s inside of 15 minutes.

    ZRM goes beyond the simple mysqldump backup I’ve been using for hicks, our production server, allowing you to configure and schedule backups by database, report on and verify backups, and restore images. Nothing that can’t be done using other tools and scripts, but it’s really quite elegant from what I’ve seen so far.

    The verdict? I’ll almost certainly be using this to protect our various MySQL instances.

  • Firefox Toolbar Replacement for Firefox:
    Not sure precisely when this was released, but an extension I’ve been waiting to see for some time was del.icio.us Bookmarks. Simply put, this rips out your existing Firefox Bookmarks system (yes, you can restore it and your bookmarks later) and replaces it w/ del.icio.us. So far I’m ambivalent toward the plugin, because it seemed to have had problems importing my Links toolbar – the one contains all my first tier links, such as webmail, Movable Type and so on – randomly dropping those items into two different categories firefox:toolbar and firefox:bookmarks. Nor is the tool terribly intuitive (what’s the difference between the Tags and Bookmarks windows?); I’m continually stumbling across random features I didn’t know existed.

    The alternative – continuing to use the archaic and vestigial Firefox bookmarks facility – seems pointless, so I’m sticking with the plugin for now. On the plus side, it’s made browsing my existing del.icio.us links exponentially faster, meaning I use the tool more.

    The verdict? I guess I recommend it, if only because keeping regular browser bookmarks seems archaic. But be prepared for a bit of a learning curve, and some installation snafus. Important note: installing the plugin will disable both the existing del.icio.us plugin as well as Google Browser Sync, though you can reenable both later if you choose.

Hope some of the above is interest. Meantime, I’ve got a couple more emails to chew through before I pack up and head for the airport.

7 comments

  1. Stephen – echo your comments about the non-intuitivity of the del.icio.us bookmarks extension, but I’ve been using it for a couple of weeks now, and would find it very difficult to go back to the FF bookmarks. I’ve found also that I am much “closer” to del.icio.us than I have been – it now seems much easier to tag/bookmark/share than before, and I think it’s just because of the ‘bookmarks’ extension.

  2. Is the link on anchortext “borderline offensive” the link you wish…? (I don’t see significant matching content there.)

    I agree that an out-of-context “web is broken” quote may not make much sense, although it may make sense in context. The “broken back button” comment has gotten my goat for years, and I’ve typed scads explaining that “back” in a navigation system may not equal “back” in an (undefined!) undo stack. That’s why I’m curious about your actual ideas, thanks.

    I’d agree with Mike, though, that knowing your audience’s capabilities is vital, when you’re asking stuff to run on their machines. Adobe Flash Player 9 for Mac/Win was released at end of June 2006, and Linux followed in about six months, as it has for about a decade… there was far more heat this time, though, because (a) rapid evolution of the Player, and the short 8-to-9 dev cycle; and (b) people couldn’t view YouTube links from their friends. The fact remains that Adobe Flash Player is the widest-support engine out on consumer machines today, and this still should be recognized by realists… FP8 hit 85% consumer viewability in 9 months, and FP9 is about 45% consumer viewability in three months. Momentous impact, that.

    I’d still like to know more about your view of offensive material, though, thanks in advance.

    jd/adobe

  3. Ric: we were just dicussing this on #redmonk today, and i’m coming around to the extension. the lack of an ability to sort the toolbar items is a killer, but it’s so much faster to tag and store as you note that i’m sold.

    John: the link is right. it’s not pointing to an actual description, just my offhand characterization of the talk as borderline characterization. this entry was intended to explain the original remark.

    “I agree that an out-of-context “web is broken” quote may not make much sense, although it may make sense in context.”

    i don’t really agree. if i were in Adobe’s shoes, i’d argue instead that aspects of the web are broken. tough to show video using straight HTML, for example, and Flash is a good solution there.

    but millions of users use the web quite effectively every day, with only small limitations. the web has been so effective, in fact, that it has forced client server applications to adapt to it rather than the other way around.

    i just don’t see how a general “the web is broken” is supportable in any context. the web needs to evolve, certainly, and is doing so, but it’s far from broken.

    “The “broken back button” comment has gotten my goat for years, and I’ve typed scads explaining that “back” in a navigation system may not equal “back” in an (undefined!) undo stack. That’s why I’m curious about your actual ideas, thanks.”

    sure. here’s what i’m talking about: first, hop over here – to the Adobe Showcase page. at the present time, the #1 link should be this – an application for some new Siemens phone. now click anywhere in the application, say the “Check it Out” button on the opening page. now hit Back. you can’t, nor can i can link to either page. when you’re operating in the context of a browser, that breaks a user’s expectation.

    “Adobe Flash Player 9 for Mac/Win was released at end of June 2006, and Linux followed in about six months, as it has for about a decade… there was far more heat this time, though, because (a) rapid evolution of the Player, and the short 8-to-9 dev cycle; and (b) people couldn’t view YouTube links from their friends.”

    this doesn’t take into account that Adobe never shipped a version of Flash 8 for Linux, waiting two full years for 9. users, such as myself, were thus unable to access content sites such as ESPN or the Boston Globe that shifted their Flash content over to the new medium.

    “The fact remains that Adobe Flash Player is the widest-support engine out on consumer machines today, and this still should be recognized by realists… FP8 hit 85% consumer viewability in 9 months, and FP9 is about 45% consumer viewability in three months. Momentous impact, that.”

    all true.

    “I’d still like to know more about your view of offensive material, though, thanks in advance.”

    let’s accept as a given that Adobe is a tremendously important firm for the future of the web. let’s also accept that Flash is hugely ubiquitous. let’s go even further and say that Flash allows designers to do things that either can’t do or can’t easily do on the web.

    that’s all good. what i don’t understand is why Adobe cannot be content to be an important part of an evolution of the web, rather than The Solution to something i don’t feel is broken. further, i don’t think it’s terribly productive to be told that technologies that work very well for a variety of applications that i use on a daily basis (e.g. Javascript in Gmail, Zimbra, Google Reader, etc) “suck” or that problems that i have with the Flash user experience are “myths.” i find that, to be candid, rather insulting.

    does that explain it adequately?

  4. Thanks, Stephen. I’m with you on preferring phrases like “aspects of the Web are broken” to “the Web is broken” — I know that lots of folks prefer the latter for its drama and impact, but I agree with you that such “labelA ‘is’ labelB” discussions often end up in strange places.

    For “back button”, thanks for the expansion… you’re looking at navigation *within* a presentation, to be controlled by the page-nav system in the document browser. Some SWF files do interact with the browsers’ document-navigation commands like that, but it’s a content-maker’s decision, rather than something built into the rendering engine. (Key search reference: “externalInterface”.)

    Me, I know the browser’s back button will go back to a previous page, but absent specific instructions on an app or preso I’m not sure what it “should” do within any particular page. I’d be inclined to put the preso’s nav system in the preso itself and remove ambiguity… hard UI problem either way.

    “Adobe never shipped a version of Flash 8 for Linux”

    That’s true. Wikipedia has a timeline. (and it looks accurate, too! 😉

    Macromedia Flash Player 7 shipped in Sept 2003. After that major architectural changes were started. Flash Player 8 (Aug05) added high-performance graphics and a new, tighter video codec, while the focus of Adobe Flash Player 9 (Jun06) was the Tamarin scripting improvements. At one point 8 & 9 were the same project, intended for release in the same version, but Tamarin required extra time. Linux went straight for full functionality. (You’ll note the “FP8.5” label in Tinic’s post… that’s how Flash Player 9 was named at one point.)

    Linux users were caught in a whipsaw here, agreed. The whole new high-performance Player took three years to implement, and the video codec was released to Mac/Win after two years, and meanwhile the Mac/Win update rate accelerated so mass audiences were more capable more quickly (influencing content creators’ choice), and meanwhile there was the whole explosion of casual video on the web.

    Thankfully that whipsaw is over right now. With Mozilla’s Tamarin Project the conversion of 32-bit JIT to 64-bit is open, and with the opening of the Linux flashsupport library there’s more leeway for more configurations. (See Tinic Uro for more context on the latter.)

    Once it was clear the logic engine required more time than the graphics engine the schedule was split; and once the schedule was split Linux went straight to the latest capabilities without stopping. When combined with the rapid audience uptake and the new video content, it was indeed a hard time for Linux users. I’m hopeful that the changes since then make things much better now.

    Thanks for taking the time to describe how things look to you — I appreciate it! 🙂

    jd/adobe

  5. If you use gentoo, please request a ZRM ebuild. The Gentoo project doesn’t take unsolicited submissions of ebuilds, and prefers not to have application developers maintain ebuilds, both for good reasons. That said, we at Zmanda would be happy to help out if contacted by a gentoo dev.

  6. Dustin:
    I’m not sure where you get the impression that we don’t take unsolicited submissions. They do tend to sit in Bugzilla until they find a Gentoo developer willing to either look after them directly or proxy changes from somebody else that is very knowledgeable – but that certainly does not mean we don’t take unsolicited submissions.

    Nearly all of the recent changes in app-backup/amanda have been from sgw, just committed by me.

    The last time I looked at ZRM (MySQL UC2006 iirc, maybe later that year), it certainly wasn’t ready for packaging in my opinion, so I didn’t handle it then. If it’s moved on considerably since then, I’ll certainly look at it again when I have time – so do keep the ebuild on the bug up to date – stale bugs are a lot less likely to get looked at than active ones.

    (My money is on mylvmbackup personally, but I’ve got coding invested there that made it a perfect fit for my production envs).

  7. Sorry, I didn’t phrase that very well. I meant to draw a distinction between e.g., Fedora, where you can build an RPM and say “here it is,” and you’re done; and Gentoo, where there’s a bit more trouble involved to get your ebuild into portage. Like I said, that’s for good reason — the project has high quality standards, and can’t accept just any old joe’s ebuild. 🙂

    I’ll be in touch privately about the ZRM ebuild, and I’d love to hear any thoughts about the amanda ebuild, too.

Leave a Reply

Your email address will not be published. Required fields are marked *