The flip side is that any reasonably popular site on the web is going to cost a not insignificant amount of money to maintain, and the hard reality is that nobody really wants to pay for it out of pocket. They want it for free, paid for with ads that other people see while they block them with their browser plugin.
I've said for a long time that I'd rather see something that functioned much like Facebook does (or did, rather, maybe 10 years ago) but was completely supported by subscription. Let me pay something like $5/month and in return no funny business, no stupid manipulation to get more page views, etc. Just a place I can interact with friends and family.
But nobody would pay for that, even if that's basically what they pay for it indirectly through the products they buy.
On the contrary, it costs almost nil to handle traffic with an HTML/CSS static site, maybe 3x - 10x that with personalization, transactions, social, etc. Most info sites don’t need those.
Define popular? How about a billion hits?
Figure four servers for static HTML/CSS, maybe add CloudFront or similar if heavy on images or progressive download video, and 10 servers for dynamic personalization/transactions, for a billion personalized/e-commerce hits a month.
A billion hits averages to under 400 reqs/second, which you could handle on a single Pentium running Apache on BSD in late 90’s. Make it a bell curve, account for surges, 4 modern boxes is still comfy headroom for static stuff.
If hosting content is prohibitively expensive you could be doing it wrong. Try not firing up script interpreters and ‘frameworks’ per hit, use a static site gen CMS. If dynamic, spend time nailing caching.
It's a pity you're being downvoted - but I realize it's for being off-topic. I feel you make a very important point: we became so dependent on dynamic websites that it stifles growth even if in many (not all!) cases a static version would do just fine - and these can withstand massive amounts of hits.
If you make enemies, though, you're screwed - a good protection against DDoS attacks can be terribly expensive.
People are willing to pay for good content. e.g. Wikipedia receives vast amounts of donations. Not only are people willing to pay money, they are willing to do it if even if the content is otherwise free.
The problem is that most of the time, ad-supported websites become experts in creating clickbait and lose their ability to create content that worth any money. It is the ads that are killing the ad-supported web, it is a race to the bottom.
The problem is that most of the time, ad-supported websites become experts in creating clickbait and lose their ability to create content that worth any money.
Depends. Some websites have a hybrid model where they have a free tier and a subscription tier that removes ads. I think that works well provided the free tier isn’t too invasive about the advertising.
What the internet really needs is a standardised format for “tipping” for content. I’d be happy to send a little money to a website that I visit without having to subscribe or put payment information in.
Especially true of something like the newspaper. I don’t actually like the NYT or Post as businesses enough to subscribe, but in the rare occasion they actually publish some worthwhile journalism rather than “let’s interview some idiots and see if they still like Trump” then sure I’d pay whatever an ad click is worth for that.
Part of the problem with that is finding the right mix of free content to attract users and still get them to trust that it's worth it.
It's also a challenge because with all of these modular services, it's hard not to start feeling like your bank account is getting the death by a thousand cuts routine.
Sure, it's annoying when I can't read a NYT article because I ran out of my 5 free articles. But is it enough for me to actually pay? No, I'd rather just read the HN comments as those are free and usually better than the actual content in the article.
The flip side is that any reasonably popular site on the web is going to cost a not insignificant amount of money to maintain, and the hard reality is that nobody really wants to pay for it out of pocket. They want it for free, paid for with ads that other people see while they block them with their browser plugin.
People pay reasonable amounts for good service. The NYT’s online subscriptions are booming, Netflix seems to be doing alright, and there are many others. What can’t survive is a model of asking for too much, or drowning out content with ads.
Ads are not the only method for monetizing content, they are merely the laziest.
Content makers have been using the web for over two decades, and figuring out how to do it profitably as well. There are innumerable models for doing it, but it requires putting in the leg work to actually care about how your work is supported. Sure it's seductive to think that someone working in some "noble" pursuit should be able to avoid having to do that, but that's not really true of anyone anywhere, why should it be true of content makers?
Comments
The flip side is that any reasonably popular site on the web is going to cost a not insignificant amount of money to maintain, and the hard reality is that nobody really wants to pay for it out of pocket. They want it for free, paid for with ads that other people see while they block them with their browser plugin.
I've said for a long time that I'd rather see something that functioned much like Facebook does (or did, rather, maybe 10 years ago) but was completely supported by subscription. Let me pay something like $5/month and in return no funny business, no stupid manipulation to get more page views, etc. Just a place I can interact with friends and family.
But nobody would pay for that, even if that's basically what they pay for it indirectly through the products they buy.
On the contrary, it costs almost nil to handle traffic with an HTML/CSS static site, maybe 3x - 10x that with personalization, transactions, social, etc. Most info sites don’t need those.
Define popular? How about a billion hits?
Figure four servers for static HTML/CSS, maybe add CloudFront or similar if heavy on images or progressive download video, and 10 servers for dynamic personalization/transactions, for a billion personalized/e-commerce hits a month.
A billion hits averages to under 400 reqs/second, which you could handle on a single Pentium running Apache on BSD in late 90’s. Make it a bell curve, account for surges, 4 modern boxes is still comfy headroom for static stuff.
If hosting content is prohibitively expensive you could be doing it wrong. Try not firing up script interpreters and ‘frameworks’ per hit, use a static site gen CMS. If dynamic, spend time nailing caching.
It's a pity you're being downvoted - but I realize it's for being off-topic. I feel you make a very important point: we became so dependent on dynamic websites that it stifles growth even if in many (not all!) cases a static version would do just fine - and these can withstand massive amounts of hits.
If you make enemies, though, you're screwed - a good protection against DDoS attacks can be terribly expensive.
I think you interpreted maintain too literally. For most sites you need to paid writers, designers, programmers, etc
Those all cost money.
GP had own site, said ‘information wants to be free’, I was talking about case where that gets popular.
I was not talking about media generated as placeholders for ads (sports, ‘news’, and reality television come to mind, as does click-bait).
Ah, fair enough
Apparently it doesn't cost money to produce content?
People are willing to pay for good content. e.g. Wikipedia receives vast amounts of donations. Not only are people willing to pay money, they are willing to do it if even if the content is otherwise free.
The problem is that most of the time, ad-supported websites become experts in creating clickbait and lose their ability to create content that worth any money. It is the ads that are killing the ad-supported web, it is a race to the bottom.
Depends. Some websites have a hybrid model where they have a free tier and a subscription tier that removes ads. I think that works well provided the free tier isn’t too invasive about the advertising.
What the internet really needs is a standardised format for “tipping” for content. I’d be happy to send a little money to a website that I visit without having to subscribe or put payment information in.
Especially true of something like the newspaper. I don’t actually like the NYT or Post as businesses enough to subscribe, but in the rare occasion they actually publish some worthwhile journalism rather than “let’s interview some idiots and see if they still like Trump” then sure I’d pay whatever an ad click is worth for that.
Part of the problem with that is finding the right mix of free content to attract users and still get them to trust that it's worth it.
It's also a challenge because with all of these modular services, it's hard not to start feeling like your bank account is getting the death by a thousand cuts routine.
Sure, it's annoying when I can't read a NYT article because I ran out of my 5 free articles. But is it enough for me to actually pay? No, I'd rather just read the HN comments as those are free and usually better than the actual content in the article.
The flip side is that any reasonably popular site on the web is going to cost a not insignificant amount of money to maintain, and the hard reality is that nobody really wants to pay for it out of pocket. They want it for free, paid for with ads that other people see while they block them with their browser plugin.
People pay reasonable amounts for good service. The NYT’s online subscriptions are booming, Netflix seems to be doing alright, and there are many others. What can’t survive is a model of asking for too much, or drowning out content with ads.
Ads are not the only method for monetizing content, they are merely the laziest.
Content makers have been using the web for over two decades, and figuring out how to do it profitably as well. There are innumerable models for doing it, but it requires putting in the leg work to actually care about how your work is supported. Sure it's seductive to think that someone working in some "noble" pursuit should be able to avoid having to do that, but that's not really true of anyone anywhere, why should it be true of content makers?
So... you'd rather be the customer than the product? I agree!