jerluc

@CmonDev We [URX] definitely have looked into support for Windows Phone deeplinking, but found it difficult to find any examples of websites using Windows Store meta markup properly, which would be a prerequisite for us to support such apps. Can you point us in the direction of any good examples?

@rahimnathwani, just to clarify one point:

There are absolutely no "designed-for-URX tags" that we have documented. All of the meta/link tags that are documented are usable by Twitter, Facebook, Google, etc. It might not be entirely clear in our documentation unfortunately, but the "omnilink standard tags" are actually just an extension of Google's app-indexing tags, and are fully compatible with their initiative.

[full disclosure, I'm an engineer at URX]

@mikejarema, this is basically what we are doing; if a user doesn't like an app, then they won't have it installed, and it will simply fallback to the website URL in the browser.

[full disclosure, I'm an engineer at URX]

@smrtinsert, just a couple of things to clarify:

1. There is no third-party service that needs to be intergrated into the mobile application for omnilinks; you can simply prepend "http://urx.io/" in front of your normal web URL, and, assuming you have deeplinks installed on your mobile application, we will try to bring your users to the app if they have it installed.

2. It does a little more than any Android intent. It works across all platforms and degrades gracefully; you can't text a plain Android intent to a restaurant on Yelp and have an iPhone recipient view the content.

[full disclosure, I'm an engineer at URX]

I was quite intrigued at first by this concept, but was a little turned off by the fact that the Sandstorm crew offers to "Use Our Servers..." to run Sandstorm itself. IMHO, this offering basically reverses any of the real benefits of Sandstorm. I'd encourage its authors to perhaps discourage the use of their "managed instances" so as to actually convince people to stop doing exactly that with other external software services.

Though I'd generally agree with your last statement (and relish in the amazement on one's face when I show them disassembled JVM bytecode for the first time), I'd have to say that this is still quite a "special" feat whether you've written a virtual machine/compiler/interpreter before or not. I haven't had enough time with the source as of yet, but I'd wager to say that there is most likely a lot more going on than just caching already-cross-compiled classes or basic opcode mappings. Not to forget that this project looks to be quite literally implementing from the JVM spec (a la Apache Harmony).

Although I tend to agree with the author's sentiments here, there is one important distinction that ought to be made here: more features/capabilities != more code != more complexity.

In my experience developing both in net-new codebases in addition to large legacy codebases, there certainly are cases where better abstractions/paradigms equate to less code, and ultimately to less complexity. Yet at the same time, these improvements tend to yield more flexibility and capability, potentially providing additional features (sometimes entirely by chance). I find that in these such cases, it actually becomes easier to reason about what your code does, because you have come to a point where you can clearly generalize about your software.

On the flip-side, I can also greatly empathize with the author, as even in my attempts to generalize the functionality of software, there often exist ugly one-off bits of critical business logic that invariably dirty the water. In these such situations, it seems to become harder and more complex to build generic code, and you are much better off just sticking to the straightforward implementation without allowing for any of the bells-n-whistles that a more generalized approach might offer.

TL;DR--It really depends on the situation, in my opinion, but this shouldn't prevent you from trying to find a generic solution that is less complex yet offers more.

Yeah...I unfortunately ended up with the same general shape: https://identicons.github.com/jerluc.png

I completely agree with this. Perhaps expanding the idea further however, it makes sense IMHO to see a multi-staff score (this could mean either a score written for multiple parts, or as something like piano scores which typically have a staff for each hand) as parallelism.

I'm assuming you never got to the "Afterword" (perhaps due to anger or frustration?), as it is at that point that the author reveals that the entire article up to that point is the exact opposite of what he supports.

This does of course make for the argument that perhaps this style of writing is not suitable for skimming, though I'd agree with another comment on this post that after reading the part about 95% of 3,875,000 equating to approximately 1,000,000, I was definitely more perceptive that something silly was going on.

Sounds like a good idea, especially the fact that it greatly minimizes the feedback time of communicating with each other (unlike email, Linkedin, etc).