The browser as a platform

I always did, I still now, and I think I will probably always, HATE web apps! :-(

In fact, I'm always looking for ways to spend LESS time in a browser.

Guess you would throw a Chromebook out the window then ;)

Me on the other hand, I can't wait for the day that my browser is the OS.
 
Just tried it now. I went to http://www.gmail.com/ from my browser. I'm already signed in so I immediately saw the progress bar that said "Loading [email protected]." From the time I hit enter on my address bar to the time I saw my inbox: 55 seconds. On the other hand, from the time I clicked "Mozilla Thunderbird" to the time it had finished checking for new mail (I had none): 15 seconds. And while I wait for it to check for new mail I can work with the mails that are already there. That I only had to wait around 3 seconds for.

Now leave GMail open in the browser for a day and you will see how I manage email.
There's no "checking for mail" because the browser ui will update by itself when there is mail.

Then try closing the browser and opening it again a few minutes later to see how fast it loads when its been cached (I hardly ever do this because the browser is left open indefinitely)

My preference is driven in part by the fact that I leave the browser open all the time anyway :)
 
Edit: Grr. Only a Chrome extension; that's not much use. As previously mentioned, I'm looking for excuses to not have to open my browser as much. :p Will try it on my iPad and phone, though. It looks quite interesting.

It's an iOS and Android app. So if a client side app uses HTTP to retrieve web site data, is it a browser platform?
 
Now leave GMail open in the browser for a day and you will see how I manage email.
There's no "checking for mail" because the browser ui will update by itself when there is mail.

So? Thunderbird does the same thing. I think it's GMail pushing the mail to me, but I can't be sure. If it is me polling, then I'm polling so fast I don't notice it.

Then try closing the browser and opening it again a few minutes later to see how fast it loads when its been cached (I hardly ever do this because the browser is left open indefinitely)

Nope. Just tried again. It takes a crapload of time to load. Maybe there's something wrong with my browser configuration?

And then I click on another mail folder (or "label") and it says "Loading" at the top for another 20 seconds or so before showing me the contents of that folder. Plus it doesn't put the hourglass on or anything. I have tunnel vision, which means I can only see the point on the screen where my eyes are actively pointed at. This means I click around for awhile before I notice that it's loading something.

Thunderbird, naturally, "just works" when I click on a folder. It takes absolutely no time whatsoever. And it would've done so even if I had no (or practically none, because it's so slow) Internet connection. And it wouldn't have cost me any money or used any of my data bundle (if I had been connecting from 3G or a capped ADSL connection, which I'm thankfully not right now).

My preference is driven in part by the fact that I leave the browser open all the time anyway :)

I get that, but even so. Now a new mail comes in. Click on it to read it. It's obviously going to take longer because it's HTML. It's not only downloading the message envelope and content (which granted Thunderbird does anyway because it's IMAP), it also needs to download all those extra tags explaining to the browser how to display it! Why? Because a browser was not designed to display e-mails! An native application built for the express purpose of displaying e-mail already knows how--it just needs to know what!
 
It's an iOS and Android app. So if a client side app uses HTTP to retrieve web site data, is it a browser platform?

Technically yes. I will use one over a generic browser (it at least has some knowledge of how to display the content, if only a little more than a generic browser), but I would obviously prefer a fully native application that rather displays the info in a grid, with actual buttons or menu items on a form, so it doesn't have to download all that extra crap.
 
The internet connection dies and your device is useless...

...if you're using a web app.

If you're using a dedicated app with decent offline capabilities (like most modern mail clients), you still have access to your historical data, and can compose new items to be uploaded when the connection finally does come back/you make other arrangements.

Sure, Gmail on the web has some rudimentary offline capabilities but (and correct me if I'm wrong here), they only work if the site happened to be open at the time your connection went down. And you can't compose new e-mails for later sending if you have no Internet connection, can you? (serious question; I actually don't know if you can)

Edit: case in point: my browser timed out after I posted this reply. I had to kill some downloads that were running and kick my wife off the Internet, and re-open the thread, hoping and praying that the message actually posted before the connection died. I wasn't about to re-type it if it hadn't. If I'd been using a dedicated client, the post would've sat in my Outbox and I could've been safe in the knowledge that it would've sent eventually without me having to do anything.
 
Last edited:
Noc why not start PAB* ? It might do you some good ;)

*People Against Browsers.

For me to write an application that would automate 2000+ open office documents is much more of a stupid idea than to automate it to a single point which is accessed via your browser. Surely you don't want to capture 2000+ open office documents replies again after it has already been filled in.

Practical example:
It's year end and you have to have your employees do their discovery health choices, now going your route you would have to setup a template, send it out, get docs back, have someone sit and capture the results. This now excluding the limitations that a document has.

Or rather write the application in silver light, have a much more secure and less limitation on how it is presented and verified, finally when everyone is done, you can press a button that reworks all the results in one go, give you reports and exports without need of few people wasting their manhours on it.

I'd take the latter, it's a lot cleaner, better to maintain and whole proccess of pre-prep and after-prep being streamlined VS for example need to trot out a template going the doc route.
 
Or rather write the application in silver light, have a much more secure and less limitation on how it is presented and verified, finally when everyone is done, you can press a button that reworks all the results in one go, give you reports and exports without need of few people wasting their manhours on it.

Or stop being lazy and write a native Windows (I presume they're on Windows since you mentioned Silverlight. Otherwise, replace "Windows" with whatever OS they use) application that will offer even less limitation on how it's presented, be way faster for them to use, and support offline storage and data synchronisation for in case their network/Internet connection goes down.

Hell, if you're really stuck on the Silverlight-style UI, use WPF. It's Silverlight on mega-steroids!

P.S. Copying this post to the clipboard before I click Post, in case it times out again... :rolleyes:
P.P.S It's not PAB, it's PAWAWNAWBMA(WAAC)*

* People Against Web Apps When Native Apps Would Be More Appropriate (Which is Almost Always the Case) ;)
 
Last edited:
Or stop being lazy and write a native Windows (I presume they're on Windows since you mentioned Silverlight. Otherwise, replace "Windows" with whatever OS they use) application that will offer even less limitation on how it's presented, be way faster for them to use, and support offline storage and data synchronisation for in case their network/Internet connection goes down.

Hell, if you're really stuck on the Silverlight-style UI, use WPF. It's Silverlight on mega-steroids!

P.S. Copying this post to the clipboard before I click Post, in case it times out again... :rolleyes:
P.P.S It's not PAB, it's PAWAWNAWBMA(WAAC)*

* People Against Web Apps When Native Apps Would Be More Appropriate (Which is Almost Always the Case) ;)
Lol, Obviously, but still upkeeping, maintaining and supporting 2000+ pc's VS single web application rollout, I'd stick to well written silverlight application. Love or hate it, you'd have to use your browser to interact with the world.
 
Lol, Obviously, but still upkeeping, maintaining and supporting 2000+ pc's VS single web application rollout, I'd stick to well written silverlight application. Love or hate it, you'd have to use your browser to interact with the world.

True, and if they're all on the same LAN I would probably (begrudgingly, though) stick with a web app as well. If it were something that they'd be expected to use more than just once a year, or there is any possibility whatsoever that they might want to connect to your app over the public Internet (or use it offline), that would be a dealbreaker for me.

We develop mobile applications, and we have proved time and time again to our customers (both for the mobile devices and the back-office application where salespeople/managers/whatever want to login from home and all sorts of other locations) that web apps simply do not work. We have thankfully reached the point where we finally have buy-in from the powers that be at our company as well. We do write web apps too, of course, but they're really really basic affairs with the bare minimum of functionality and the smallest possible HTML to ensure the fastest possible response. They're nice in a pinch (I myself have used Outlook Web Access to check my work e-mail before I got a Smartphone because it's too much damn mission to boot up my notebook and connect to the office VPN).

Everyone always talks about the user experience. But in reality it isn't about the user experience at all, is it? It's about the company's bottom line. Charging the customer as much as we possibly can while doing the work in the shortest possible time to ensure the largest possible profit for our company. It's absolutely sickening. If it was all about the user experience, I don't believe for one minute anyone would use a web app as their primary means of working (although, as I said, they're nice in a pinch). :(
 
Lol, Obviously, but still upkeeping, maintaining and supporting 2000+ pc's VS single web application rollout, I'd stick to well written silverlight application. Love or hate it, you'd have to use your browser to interact with the world.

True, and if they're all on the same LAN I would probably (begrudgingly, though) stick with a web app as well. If it were something that they'd be expected to use more than just once a year, or there is any possibility whatsoever that they might want to connect to your app over the public Internet (or use it offline), that would be a dealbreaker for me.

We develop mobile applications, and we have proved time and time again to our customers (both for the mobile devices and the back-office application where salespeople/managers/whatever want to login from home and all sorts of other locations) that web apps simply do not work. We have thankfully reached the point where we finally have buy-in from the powers that be at our company as well. We do write web apps too, of course, but they're really really basic affairs with the bare minimum of functionality and the smallest possible HTML to ensure the fastest possible response. They're nice in a pinch (I myself have used Outlook Web Access to check my work e-mail before I got a Smartphone because it's too much damn mission to boot up my notebook and connect to the office VPN).

Everyone always talks about the user experience. But in reality it isn't about the user experience at all, is it? It's about the company's bottom line. Charging the customer as much as we possibly can while doing the work in the shortest possible time to ensure the largest possible profit for our company. It's absolutely sickening. If it was all about the user experience, I don't believe for one minute anyone would use a web app as their primary means of working (although, as I said, they're nice in a pinch). :(
 
True, and if they're all on the same LAN I would probably (begrudgingly, though) stick with a web app as well. If it were something that they'd be expected to use more than just once a year, or there is any possibility whatsoever that they might want to connect to your app over the public Internet (or use it offline), that would be a dealbreaker for me.

We develop mobile applications, and we have proved time and time again to our customers (both for the mobile devices and the back-office application where salespeople/managers/whatever want to login from home and all sorts of other locations) that web apps simply do not work. We have thankfully reached the point where we finally have buy-in from the powers that be at our company as well. We do write web apps too, of course, but they're really really basic affairs with the bare minimum of functionality and the smallest possible HTML to ensure the fastest possible response. They're nice in a pinch (I myself have used Outlook Web Access to check my work e-mail before I got a Smartphone because it's too much damn mission to boot up my notebook and connect to the office VPN).

Everyone always talks about the user experience. But in reality it isn't about the user experience at all, is it? It's about the company's bottom line. Charging the customer as much as we possibly can while doing the work in the shortest possible time to ensure the largest possible profit for our company. It's absolutely sickening. If it was all about the user experience, I don't believe for one minute anyone would use a web app as their primary means of working (although, as I said, they're nice in a pinch). :(

:eek: @double post

/checks matrix.

Well said.
 
Everyone always talks about the user experience. But in reality it isn't about the user experience at all, is it? It's about the company's bottom line. Charging the customer as much as we possibly can while doing the work in the shortest possible time to ensure the largest possible profit for our company. It's absolutely sickening. If it was all about the user experience, I don't believe for one minute anyone would use a web app as their primary means of working (although, as I said, they're nice in a pinch). :(

So you think the end-user support for a native app running on 10000 PCs each with a different configuration will be cheaper? Not a chance.

Unfortunately (or rather fortunately) you are one of a dying breed, as the world is done with thick clients. Have a look at Windows 8.

Next argument: "You can only do so much with HTML"... You have obviously not seen HTML5 and CSS 3. On that note, Adobe is getting rid of flash, as everything flash can do, can now be done with a combination of HTML5, CSS3 and JS. (See Adobe Edge).

Next argument: "It's not what the web was designed for"... Well switch your microwave, GPS and television off then, they are not what their original technologies was designed for. (Them and almost everything else we use). Thankfully we've got some human ingenuity keep coming up with new ways to use the same things.

Every PC and most devices has a web browser installed. Most of them do not have Thunderbird or OpenOffice installed. I would most definitely rather know that I can access everything I need to from anywhere without needing anything more than a user name and password. While the native apps do have a lot more functionality (sometimes), that's a sacrifice I'm willing to make.
 
Last time I used thunderbird I thought it was rather meh. Would rather stick to gmail on the web. Its quite slick really.

@Nocturne, how do you feel about Outlook web access then? :p
 
Just like tv didn't kill radio, both desktop and web apps will coexist for a long time to come.
 
Top
Sign up to the MyBroadband newsletter
X