Slashdot Mirror
Ask Slashdot: Why Do Popular Websites Add New Features So Sparingly?
dryriver writes: If you are a user of a popular professional desktop software program, it is not uncommon for that program to get anywhere from 5 to 20 major or minor new features and functions about once a year to stay desirable and competitive. But it seems that hugely popular internet-based sites and services like Instagram, Facebook, YouTube, Google Search, Gmail, Outlook, WhatsApp, Telegram and others get major new features/changes much, much slower than desktop software. Quite often you'll come across a barrage of breathless news articles that say "Popular Internet Service X will add Y feature starting from April 1st." It is often one single and very obvious feature or functionality being added that people have wanted for years, not a cluster of 5 or 10 funky new functions at the same time.
Why is this the case? How is it that desktop software with just a few hundred thousand users and no more than a few dozen coders working can add 5 to 20 major new functions in just one year, and do this year after year, but a major internet-based service with tens or hundreds of millions of users and presumably hundreds or thousands of techies working behind the curtain keeps everyone waiting three years or longer to build a much requested feature into the system, and then only rolls out that one desired feature to great fanfare as if it is a huge achievement? Is it really that much harder to code major new features into an internet/cloud service, versus coding major new features into desktop software; or is this a deliberate business model that has become popular?
Why is this the case? How is it that desktop software with just a few hundred thousand users and no more than a few dozen coders working can add 5 to 20 major new functions in just one year, and do this year after year, but a major internet-based service with tens or hundreds of millions of users and presumably hundreds or thousands of techies working behind the curtain keeps everyone waiting three years or longer to build a much requested feature into the system, and then only rolls out that one desired feature to great fanfare as if it is a huge achievement? Is it really that much harder to code major new features into an internet/cloud service, versus coding major new features into desktop software; or is this a deliberate business model that has become popular?
We see right through this one, slashdot. You haven't added features in a decade or more, but that doesn't mean that this site is popular or relevant because of it.
Damn_registrars has no butt-hole. Damn_registrars has no use for a butt-hole.
Don't fuck with it.
whole bunch of people need to learn that...
By the time a web developer has learned the latest JavaScript framework, a new framework has already came out and the old framework is being depreciated.
Goodbye, Slashdot!
1. The software is mature, and any "new features" are just Gold Plating https://en.wikipedia.org/wiki/Gold_plating_(software_engineering) (See MS Office suite).
2. The software is NOT mature, but any new features become an arguing match between Developers, Marketing, Upper Management, etc, so thus only minimal changes are ever made. (See Facebook)
3. The software is old, krusty, and incredibly hard to maintain. Adding anything new that would truly be useful is a gargantuan task in painful software archeology. (See Slashdot)
A major service website (like the ones listed in TFS) is defined by its basic function. Facebook provides communications between users. Google is a search engine, Outlook is a mail program, YouTube shows videos. Once the major function of the website is defined and accepted, adding new features and functionality will be confusing and offputting to the users.
Applications, on the other hand, must support new types of data, new data locations (ie cloud services), different display and printing options and etc. In terms of continually updating applications is for some vendors (*cough* Microsoft *cough*) is a source of revenue.
When you talk about why are there lots of coders for websites versus few for Applications, I would point out that you aren't looking behind the scenes at a website - many coders are required to implement new technology to bring the services faster and more reliably to more users as well as keeping ahead of the bad guys.
Mimetics Inc. Twitter
Following every hype, adopting features fast and without clear goal, etc. is called "bad engineering", incidentally. The problem is that there are a lot of bad and really bad people at work on the web and on apps that I will refrain from calling "engineers" because they do not deserve that title. Hence doing it right for a change stands out. In other engineering disciplines it would not or at least not nearly as much.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
That's like asking why books rarely add new chapters.
Desktop apps need to continually add features to attract new customers, since they get the bulk of their revenue from the initial sale, or perhaps from an ongoing subscription. So they are in a constant arms race with their competitors, and need to improve lest they are overtaken.
But Facebook, Google, Youtube, Gmail, etc. are advertising driven. Every person using their platform represents some amount of recurring revenue, so continued user growth isn't as big a concern until they actually start losing market share.
But I do question your basic premise of user populations and developer counts. Certainly Google and Facebook have literally billions of users, but Windows, Office, Photoshop, Steam, etc. have very large userbases and large development teams.
To Wit: /. Beta. Did not want, do not want, what is now is fine... just fix the goddamned unicode problem.
Seriously. The quest for the New Shiny more often than not just ruins things. Like round picture frames in contact lists, etc. Who wants that?! Square was just fine. And flat UI designs.. they universally look like something a preschooler did with safety scissors, brightly-colored construction paper and paste.
The "Civilized World" jumped the shark ca. 1973.
Websites market to basically every human. That means the 95% of humanity that isn't tech nerds. There are a few other software suites like this. Namely Office Suites. MS Office hasn't really had a big change since 2007 -and LOTS of people hated it when they did. Facebook, Google, etc. all have to cater to the bottom 90% of users. That bottom 90% doesn't like change very much, so features are added very slowly.
Most all the rest of the software on the planet is marketed to tech nerds -people that will actually PAY for software. To keep the money coming, the companies have to keep new features coming.
I think the question answers itself in the second paragraph.
"How is it that desktop software with just a few hundred thousand users and no more than a few dozen coders working can add 5 to 20 major new functions in just one year, and do this year after year, but a major internet-based service with tens or hundreds of millions of users..."
Small software projects need to attract users, websites with millions of users don't, they already have full userbases. Small projects only risk angering a few thousand people, big websites will tick off millions with even the smallest change. It's simple economics.
Besides, a desktop application probably only needs to support a couple of operating system versions. Websites need to work across dozens of devices (phones, browsers and hundreds of combinations of both).
The question is kind of like asking why an aircraft carrier costs more than a VW, but can't turn as sharply.
.... is software that DOESN'T add new functionality all the time "Popular", and software that DOES adds crap all the time not so popular?
As a software developer myself, my gut feeling is that you have somewhat of a sweet spot with a 20:1 User:Developer Ratio, and maybe 200 users and 5 developers for getting "new features" out quick.
When you reach hundreds or even millions of users, the amount of work needed to define and support the new features (both technical and for user testing/training) becomes so great that it becomes pretty much impossible to add new features. Or at least no longer profitable.
...the desktop software is the product, and thus needs to be upgraded for the revenue stream to keep up.
For all of the web sites cited, YOU'RE the product, and you can't be upgraded.
... are designed by idiots trying to increase ad space. Let's be honest. The new site redesign at reddit is much worse and less readable then old reddit. The reality is if reddit and other sites want more ad space they'd do well to create a completely seperate site from the main reddit site.
Most enhancements to the user interface are designed by total idiots. It's not that "innovation" is bad, it's that you need to think about the person using the website instead of business focus based bullshit. Many of the reasons people use social media like reddit or slashdot is because they got the user interface design mostly correct even if there is some cheap or bad design.
Instead of saying "how do we expand our audience or our reach to make more $$$" try to understand perhaps you need to find other avenues of making money besides selling ads or transforming a website from why any group of people found it interesting in the first place.
Don't fuck around with my user experience. If I see something I think would help, I'll ask for it. If I didn't go looking for a solution, there wasn't a problem.
Websites don't push out new features? They do all the time. What they don't do is announce them- they tend to just roll them out. So you get a constant barrage of small updates. Facebook in particular- I worked there. It gets new features daily. To the point where the people working there don't even know what's going out- whenever discussion of making a "what's new" type announcement was brought up, they basically decided it was impossible to keep track. If anything your description is precisely backwards.
I still have more fans than freaks. WTF is wrong with you people?
Hey, let's take something that works and mangle it till it is useless!
It's the business model.
A service website typically gets revenue (directly or typically indirectly) from use. Once it's working, popular, and supporting most potetial customers, the bux roll in. Why change what's working and risk breaking that? Essentially only bug fixing and reach-expanding could pay for itself.
An application typically gets its revenue from sales. Once it's sold, the user has it. No more money from him. Given time you saturate the market and your revenue peters out - while your support load continues.
This can be fixed partly by making the app run on other platforms and expanding the target market. But for ongoing revenue you need previous customers to buy again. They won't do this unless you provide a later-and-greater version with enough extra functionality to be worth it. Then they're in the business of adding bells and whistles until the old customers become repeat customers.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Adding features for a (hopefully customer-driven) purpose ***can*** be a good thing.
Adding 'features' just 'cause you can (e.g. microsize grey-on-grey text; irrelevant hipster-bait pictures, widgets. and fluff that send the information density to limit-case zero...) SUX massively.
“Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.”
Antoine de Saint-Exupéry, Airman's Odyssey
Of course this philosophy CAN become pathological (Macbook; Gnome desktop)...
Desktop applications only have to worry about one user, Websites have to worry about all of them. As a result, in the early life of a website (with few users), it's relatively easy to have engineering focus on features, as most available web tech these days can handle that. As your user base grows, however, you start running into scale issues where features you've previously built don't hold up so well. Suddenly, you're putting a good chunk of your engineering effort into updating your existing features for the new # of users required. At the same time, the effort of adding a new feature becomes harder, because you can't just create a new one like before - you have to engineer it to perform at the scale of your current (and future) system, with all the yak-shaving of technical debt which may be involved.
In short: As things get bigger, they get harder.
This signature can save you $400 on your car insurance!
I wish Facebook and Google would completely overhaul their UI every few years, to remain 'cutting edge'. I mean, the KISS principle is so last century! /s
And even at that slow rate, they keep adding things to web pages that make them worse rather than better.
Beware of Sales Reps bearing gifts.
I guess you weren't here for the last redesign.
After 3 years of beta testing and consistently being panned by users, the new owners tried to force it onto users, it went so bad that users staged a boycott to get it rolled back.
a desktop app rolls out on a per-user basis, over time, and the desktop app is behind a users' own firewall (or NAT), which is their responsibility to maintain.
a webapp: you make one deployment error or one security snafu and the entire userbase - your entire business - comes crashing to a halt, effective immediate.
there's really no comparison. running a web service is scarily unforgiving of mistakes. the only real way for this to be fixed is to change the paradigm of what constitutes a web app: distributed services, distributed databases, and the web "app" be downloaded by the user and under the user's control as to when and whether they "upgrade". which requires such a large paradigm shift as to make it extremely difficult to consider. the only company that can be said to have successfully deployed this paradigm is google, with ChromeOS.
For software, the income comes from the people who pay to get the software. No new features means no sales, since no new features means there's no need to upgrade to a new version.
For Facebook/etc, the income comes from the ads pushed to the users. Too many new features at once and too much difference between versions and your risk loser your precious users.
Next question?
#DeleteFacebook
The bigger you are, the more it costs. If you're a popular service, you're likely running many, many servers doing many, many things across a broad geographic area. You've carefully implemented your infrastructure to balance cost, stability, reliability, and performance. Adding just one new feature can completely upend this calculus. If you're running multiple server farms in multiple data centers, this gets expensive quickly.
You can't afford "aw shucks oopsie woos". Whoops! Your new feature caused some unexpected behavior for 15% of all users, resulting in 18 hours of downtime! If you're a small web operation, you're sending out a lighthearted email apologizing for the inconvenience and promising to do better. Maybe you're even offering a week's worth of free service. If you're a major player, you're in the world news. Your enterprise customers are screaming at you--or worse, they're not screaming at you and are looking for your replacement. You're working on figuring out just how much this will impact the bottom line, because if you're going to need to cut back somewhere, you want to know that as early as possible. Mess up hard enough, and you're looking at a subpoena from your governmental bodies of choice.
You can't afford to annoy your users. Ooooh, we've all had that time when we rolled out an awesome new feature and the user response ranged from "meh" to "change it back right now you gibbering twits." That's never fun, is it? Gotta roll back to yesterday's configuration, apologize, and try to figure out how to move forward. If you're a major player, "rolling back" may be nigh impossible, and if you've already reconfigured your infrastructure to accommodate your new feature, that's money already spent (and worse, your new configuration may even be sub-optimal in the absence of said new feature.) You're basically looking at the same outcome as the previous point, perhaps minus the subpoenas and plus a bit more global mockery on social media.
Messing up will cost you users, and those users are unlikely to return. If you're small, this can be weathered, and is almost expected. There are way more fish in the sea, and you if can iron things out, you've still got plenty of room to grow. If you're big, everyone already knows about you and what you do. You've got a lot smaller pool of "new" people to bring on compared to the people you've already reached. Big companies that mess up need to work to retain unhappy customers, because there aren't that many fish in the sea who haven't already heard of them.
Obliteracy: Words with explosions
Why Do Popular Websites Add New Features So Sparingly?
If it ain't broke, don't break it.
It must have been something you assimilated. . . .
We need examples to be given of this desktop software that has 'a few hundred thousand users' that features are added to ever year.
Just a few examples of software would be helpful. I can't think of anything with those sorts of numbers, myself. Much desktop software either has a huge userbase or is tiny.
... I don't want a reward or nothing.
It little behooves the best of us to comment on the rest of us.
Since a lot of desktop software is sold via a license, they need to make a big deal out of the "shiny new version" in order to get your upgrade business.
For websites, advertising is the model so sites just quietly add features all the time to keep current - and since it is continuous they don't make a big deal of it.
New features aren't always good. They tend to encourage bloat-ware. I remember when Nero BurningROM was the best burner software out there (IMHO).
They kept adding more and more features and more and more bloat. A 10MB download (for the installer) turned into a 20MB, which then turned into a 50MB download, which (last I checked) had topped out around 117MB...
For a fucking burner?!?
I stopped using NERO a long time ago... It became JUNK.
I wish, sometimes, that Windows apps had the UNIX philosophy of "Do 1 thing and do it well". But, you can't sell the exact same app to a person twice if nothing has changed, so I understand WHY it happens, but it eventually kills most payware..
Wow, I think the opposite is true. Everyone and their dog seems to be adding endless new JS crap to websites that no one needs.
"What lies behind us, and what lies before us are tiny matters compared to what lies within us." Ralph Waldo Emerson
Split your examples into major services, and small specific services, then compare them to desktop software. I mean properly!
Facebook? Features get added, removed subtle UI changes etc on a monthly basis. Google does so on an event driven basis (Have you been watching the world cup ticker updating live in your google results?) Did you notice the change to maps voice navigation that rolled out last month in how it announces locations? Did you see the added feature that asks you to confirm traffic accidents?
Of course not. You're not paying attention.
Likewise how many value added features were constantly changing on ICQ? What has changed in Skype other than the number of adverts that are shown? Yeah they move the buttons around but features? At the start of this year they added the ability to on the fly switch between multiple cameras. Back in 2015 they added group chat.
If you have small apps, websites, or special purpose programs you don't need new features regardless of the platform.
If you have large software or web apps you will get new features regardless of the platform.
Pay attention.
I was expecting more mentions about digg, it's such a famous example
Slashdot Beta is exactly what I thought of. Then they rolled out their really crappy "mobile" version. Fortunately they have the "use classic" link so we can keep using the design that works well.
Most of the time, you can kill automatic updates by adding a hosts file entry setting updates.steam.com or whatever to 127.0.0.1. You have to find the right hostname for each software you want to block updates on.
I'm 90% sure that was from an APK impersonator, but it was still damned funny. If I could I'd give it a +1 for that.
Damn_registrars has no butt-hole. Damn_registrars has no use for a butt-hole.
The versions numbers of OS, web browsers and what is supported.
They know how many use add ons that will not support other non https networks and block data from another domain.
That needs a https connection to one trusted domain. Bandwidth costs for a huge site load per view becomes a matter of profit.
Keep it simple and fast with less data costs that support all computers OS and browsers expected to view the site.
Domestic spying is now "Benign Information Gathering"
I work for Google Search. We launch multiple new features daily. Thing is, you don't necessarily see them all.
Some changes affect ranking; improving the algorithm, bringing you even better results, coping with misbehaving websites, and more. You won't see any UI changes here, but the search results get better. This could affect all of search (rare, but happens), or queries of a specific domain (e.g., queries about music), or a specific subset of results (e.g., sites that don't use https).
Some changes affect performance. A change that shaves a dozen milliseconds off the result page's load-time isn't something you'll notice, but in aggregate, these make Google Search better. Again, some of these optimizations apply to all searches, some to subclasses of searches, or to some devices only.
Some changes make our direct answers better. You're probably familiar with Google's calculator, for example; how many people would note if it suddenly starts answering queries involving a few more units, or different ways of asking about math? Google also provides answers about sports, weather, stock quotes, movie showtimes, and more. A new feature could involve better weather predictions, or supporting new leagues worldwide, or having fresher results. (If you're following the soccer World Cup, try searching for that on Google.)
Some changes involve only specific languages, or specific platforms (only desktop, or only mobile, or only iOS, ...). Some changes are experimental, and are removed after a while if they don't prove popular with users, or ephemeral, and are removed after a while because they're no longer relevant.
In short... to say that Google rolls out few and far between updates is somewhat inaccurate, and I'm sure the same is true for all the other websites you mention in your question.
[I work for Google but this response represents only my personal opinion and is not official in any way or manner.]
- Tal Cohen
So, full disclosure, Nero gives me free beta/release copies for review and such. Even ignoring that, I feel the need to come to their defense just a bit, because it directly relates to the feature-bloat balance problem.
Yes, Nero's code got big. Far bigger than maybe it needs to be. However, Nero quickly found itself in a no-win situation.
In 1999, everybody with a CD burner had Adaptec CD Creator, with Nero being a purchased alternative or, in some cases, bundled with aftermarket burners. Either way, everybody was burning CDs, and paying $70 for a CD burning application made sense. Nero 6 will always be my favorite version as it was arguably the highlight of the title. However, as the iPod started to gain traction, high-capacity external hard disks that could sustain 40MBytes/sec over USB 2.0 became affordable, broadband started to make things like Carbonite viable, the need for dedicated CD burning software started to dwindle. Aside from Discspan's ability to auto-sort data to optimize multi-CD burns release in 2016 (I think), there really wasn't much to add to Burning ROM between v6 and 2016. I use Burning ROM when I need to burn discs because I happen to have a copy, but can I name the last time my CD burning needs weren't adequately met by InfraRecorder or IMGBurn? Off the top of my head...not really.
Nero opted to branch out into tangentially related media creation software, ultimately settling on making a consumer video editing title, along with a media streaming/library software that takes aim at Plex. Yes, it's big...but compared to Cyberlink, WinDVD, or Adobe, it's the smallest.
All that being said, its size is largely due to the shift in function. Sure, Nero isn't as ubiquitous as it once was, but do you know anyone still using Roxio? I doubt it.
With straight audio/data disc burning adequately covered by OSS and any version of Nero released since the Bush administration, the result is code bloat. The same happened to MS Office (can you name more than half a dozen useful new features since Office 2003?), iTunes, Photoshop, Winzip, PowerDVD, AutoCAD...the list of software in the same boat is quite extensive. Nero shifted focus. Microsoft went to subscriptions.
People typically don't like when their environment changes unexpectedly.
So, if you make substantial changes of popular site, you unavoidable get your old users angry.
Even if they need these features (and typically they are not - they use just a small subset of existing features), they may think that need to change their habits is too big cost for new feature.
If you produce desktop programs, your users have an option to continue use old version.
Lot of people still use Windows XP ten years after new and shiny version of Windows appear and after few years of gratis upgrade campaign.
It is pain in the ass to support several major versions at once, but vendors have to do so. Because users don't upgrade until something forces them to. And if "something" is policy of sofware vendor, you have big chances, that user would go to competitor. One have to relearn software anyway, when upgrading. Why not learn other product, which promises not to force such a trouble on you.
It is almost impossible to keep several versions of website running on the same host and let user choose look and feel in their preferences.
So, if you want to attract new users, you can use shining new features. If you want to keep your old users, you should be as much conservative as possible.
Remember - progress is evil.It is just too often appears to be lesser evil of all possible choises.
More like big vs small. Excel 2016 is still unable to open 2 files with the same name...the difference between major version of big desktop software is usually very small, they just add a version number (and possibly some L&F changes) to look new and convince people to upgrade. Also, new releases split by months/years. Small/New software and websites add features way more often. Telegram was a bad example of a website in TFS, since it adds new features very often hence invalidating the point. Well of course websites have to be a bit more cautious since updates will affect your entire user base immediately, limiting features to a group of people is more difficult (e.g. beta software) and I guess not always doable.
That's why I still use the page with the link:
"https://slashdot.org/?nobeta=1"
Web developers are constantly working on the most requested feature: speed. That's not a problem for most PC programs, which can access data with a much wider bandwidth, run on powerful CPUs/GPUs (do not have to deal with cell phones), and do not have to work around the bugs of a thousand different browsers.
Then how come websites slow down over time? I can understand limited performance if you're trying to run an interactive application through interpreted web languages, but it seems as time goes on, pages that should be serving basically static content (news articles, blogs), slow down.
How much cash do you get for letting the voices in your head type? Seems like the return on that must be pretty poor. It's not like anyone ever paid for a hosts file manager.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
The advertisers and data buyers are who get majority of upgrades and updates you cannot see. You are the product.
A bit like pigs complaining that the slop they eat is always the same. As long as it flattens them up then fit for purpose.
What youtube has very publically fucked with its site repeatedly over the last year or so adding unwanted redesigns and functionality. Their latest idiot functionality is to make the subscription list page less useful by applying some dipshit algorithm to show interesting content. Never mind they have at least two other pages that do that.
Their other recent brainwave is to kill off Vevo and other artist pages by combining several distinct accounts into one "youtube official" feed per artist and removing the distinct ones from the users subscriptions. This may be appropriate where the user doesn't subscribe to all the various "official" accounts, but is entirely pointless otherwise, and breaks the fundamental concept - don't fuck with a users subscriptions.
You'll notice their idiot updates have been against users who don't use the site the way they want them to use the site.
If it ain't broke, don't fuck with it!
Something a lot of sites & software companies coulda/shoulda learnt by now...
Amazon releases an update, on average, every second.
Google updates its search algorithm about twice a day.
Do you have a Windows computer? Microsoft updates the OS and many of its desktop applications on a weekly schedule through Windows updates.
Release frequency is not related to desktop vs. Web. It's related to the budget a company has for software development.
Whenever a change is made, the metrics show users have difficulty with the change. Even if it doesn't, you have to suss out the effect, which gets fuzzier with more changes. Over time, marketeers decide the less change the better.
Some people really hate change. They are also very vocal. /dev from the old days.
Take the 386 chip. It wasn't good when it was good. 68000, 88000, other chips were far superior to the 386. Yet we still have the sucker around today. Even when it was painless as it would have been with the Digital Alpha chip. The first 64 bit CPU, pre-dating Apple by years even though Apple lied about it when they released their chip.
Slashdot beta - lots of complaining, even on this discussion.
Systemd - We needed a new init, man the bitching about this one. So many didn't want to suck it up and learn the new system. At least there wasn't as much bitching about fixing
And so on.
They should teach kids in school. Things change. Has it been tried before? If it was tried before was it successful? If it wasn't successful, like socialism for example what will be different this time? If there is no change, it'll fail just like it always has.
Hey, let's take something that works and mangle it till it is useless!
You know I adjusted just fine 10 years ago with Office 2007 after 1 week. I can't live without it now as I am used to hitting the shift key for the keyboard shortcuts and I can now use Word without a mouse on a plan because of this.
No more unproductive nested menus. Sometimes change is hard and you need to use your muscle memory to learn new things.
http://saveie6.com/
Hard to add stuff without screwing up what's already there.
deleting the extra space after periods so i can stay relevant, yeah.