If you want to be in control, the solution has always been the same: own your own domain, and post there.
If you post on someone else's domain, then it's their content, and they're in control of it, not you. No matter what is claimed otherwise.
Yes, it's harder. Getting visibility is also harder. It's still the only solution if you want to be in control.
It's okay to pay someone to run the blogging platform, etc., as long as you own the domain name & can decide what it points to. You can then switch providers, switch systems you use, or whatever you want to do. But if you don't control the domain name, someone else is in control.
This issue is why I’m skeptical of SaaS in general.
I look at software as a solution to a problem. When I have a problem and software solves it, I’m happy. The better the software, the more thoroughly it solves my problem without creating new problems for me (by being annoying or inefficient). The best software seems to hit a sweet spot with the right set of features targeted at solving a particular problem or small class of problems with minimal fuss.
Really good software becomes extremely difficult to improve on. But services can’t deal with just hitting the sweet spot and calling it a day. They need to grow, which means they need to evolve. The constant cash flow seems to burn a hole in the manager’s pocket. Eventually they cross a threshold and exit the sweet spot. Maybe they release a major version with an awful new UI. Like a TV series which has gone on too long, they jump the shark [1].
Perhaps this problem of SaaS may be analogous to network decay [2], though the latter is more gradual.
1) Control - SaaS usually ingest my data and silo the results of my work, so that I don't own the output. This is often unacceptable. It also makes solving problems harder. With a local application, I can always hack around its deficiencies, or easily post-process its output. Can't do that with a webapp owned by someone else.
2) Reliability - While a SaaS will be more incentivized to support a paying subscriber, pay-once software with no auto-updates is more reliable once it solves your problem. That's because it's frozen in time. It doesn't get new features (which usually aren't needed if you have a working solution already), but it also doesn't break. In the context of your whole computing system, pinning to a version of a local piece of software lets you freeze one component of your system - which reduces complexity, and gives you one less thing to worry about.
(Longer term you may need to worry about the rest of your system becoming incompatible with the version-pinned program - but VMs essentially solve that these days.)
3) Relationships - Every SaaS I subscribe to is another relationship I enter with a third party that needs to be consciously tracked. I don't want that, it's anxiety-inducing and cognitively exhausting in the long run. What I want is to pay money in exchange for solving problems. I don't want new relationships on top of that. I tend to use the following example: when I go out to buy some bread in the morning, I exchange money for bread, and that's it. I don't enter into an ongoing relationship with the store, the bakery, the flour supplier or the wheat farmer. I like my software like my bread.
(To be more accurate: I actually do enter into a bunch of relationships when I buy food or trinkets in stores, but they're all governed and fully handled by various business and consumer protection laws - which means that I don't have to care about them. They only matter if something goes wrong with the goods I bought - and then all I need is to keep a proof of purchase and know who to e-mail about it, and I can even CC the government to force-handle things for me. Meanwhile, SaaS creates relationships you do have to track yourself.)
While I share the frustration with the constant need to fiddle with things, I am not sure if going a non-SaaS route is gonna solve this problem completely. Pay-once software at least has a clearer incentive to deliver improvements, so that you want to pay for the newer versions. So the need to grow might be even stronger. Ultimately old versions will also go out of support.
Maybe OSS is the best bet, as long as the project is popular enough.
SaaS software on the other hand is more popular since the revenue stream is forcefully consistent no matter if you're putting out 10 features in a quarter or 1, or if the customer even wants to pay for whatever new feature the SaaS is putting out. With pay-once software, your revenue stream will fluctuate more depending on if you're putting out features everyone wants or not, which is great as a consumer (assuming your needs match with other customers) but could lead to huge financial issues (read: layoffs) if someone makes the wrong decision and develops features for 2 quarters in a row that no existing customer upgrades for.
"This issue is why I’m skeptical of SaaS in general"
SaaS is eating the B2B (Busines to Business) software market. Companies small or large are much happier not having to manage servers, installation, updates and security.
I don't particularly like SaaS either, but what is the alternative? It is impossible to self-install a web app on a server in a way that is easy and simple to do particularly for non-techical users. Until that happens (which may be never), SaaS will continue to grow and dominate the B2B market.
Imagine if installing a web server app was as easy as installing a desktop app. It would unlock countless opportunities for developers. My impression is that developers think it's a non-issue or have simply never thought about it.
(And no, before someone mentions it, Docker, Sandstorm, Cloudron, command-line installations etc are not easy-to-use options for non-technical users.)
Very hard. Terrifying to me as a professional developer. I can't keep up with all of the possible failure paths. The more of that I can foist off on somebody who does it full time, the better. Then I can focus on my application domain rather than the basic mechanics of keeping the thing up and secure.
For installation, install apps feature on NAS product like Synology (that internally uses Docker) is easier. Managing, keep updating, HA, etc is different story.
Note this is not specifically SaaS-related. In more traditional models, you also have new releases and upgrades, therefore you need to evolve in order to convince your existing customers the upgrade is worth the cost. Sometimes the new product is a failure (Vista, Windows 8), sometimes they see the benefit and will migrate on their own (Windows XP, 7). In any case, you are forced by the marked to endlessly add new features to stay competitive. In practice, many of these features just use up precious system resources and create problems.
Even Free Software is vulnerable to this problem - Gnome is a good example. But fortunately in these cases people can take things into their own hands, and sometimes they do - and eventually we have such nice projects like Mate and Cinnamon.
Exactly this. And if you have a problem, you don't want to go back to it every time the service changes fundamentally, they go out of business, get acquires by someone etc.
So much this. It happened to me when 1Password iPhone App changed and no longer required you to enter your master password with every phone restart. And that same month I made a minor tweak to my password (because I was using it so much for a new computer) and got locked out.
Other zero-value improvement was with Transit iPhone App. They made cosmetic GUI changes which make no sense (cover 50% of the map, swipe interaction target too big, transit line notifications screen gone) and they made zero improvements to transfer scheduling.
Weather Underground iPhone App took a compact and information rich GUI and turned it into a garish nightmare with 'speedometer' temperature graphic.
I've said it here again and again--I don't understand the point of ruining a good iPhone App. Why not treat it like a video game--just make a new one and support the old one with security updates? Heck. I get it. Double the code base, without the _perceived_ possibility of a revenue boost from the _new and improved_. At the very least make improvements very carefully (1Password), and if you have a hit don't redesign. Maybe --- make the old version the subscription and the new software ad-based? Maybe then you'd at least have a real measure of _new and improved_.
> But services can’t deal with just hitting the sweet spot and calling it a day.
The problem with the free market is that "sweet spot" is usually defined as what the majority wants. That's why I think we need open-source, so small companies can deal with the long tail.
Good advice, but this rant is about the tool not the domain ownership though.
You can own your own domain and use Wordpress self-hosted, then find they updated the UI constantly as you upgrade it keep it up to date for - you know - security.
The only way to get the level of control long term is probably sadly a static site generation kind of a thing.
There was a HN post about "software that is no longer changed but still used" as I am paraphrasing and this is an interesting idea.
A GUI blogging platform that ends it's life and stays as-is forever would be cool, even if it lack funky features, maybe you enter the HTML yourself.
I agree, you can own your own domain name without having to do everything yourself. Like anything else, if you don't want to do something, hire somebody else to do it.
I own my own domain name, and then pay others to take care of my website. I chose a cheap one for now. If I don't like their service, I can move my content somewhere else in the future. I could even have it hosted for free on GitHub, if I'm willing to live with him there constraints on static servers.
But I own the domain name, so no one can just take it down without any recourse. If someone decides to stop providing me a service, I can just take my stuff somewhere else.
Of course, you have to have a backup of the data, and use tools that are available elsewhere. But that really isn't that difficult. Static sites are really easy to manage this way. If you want something beyond the static site, it's not hard to use tools like WordPress which are open source software and widely available elsewhere. (Be cautious about adding plugins to Wordpress, some of them are security disasters, but at least you get to choose what you use.)
If you want your services to be in control of you, self-host everything. That way all of your weekends and nights are for updating to FreeBSD 14 and debugging why postgres can't write to disk anymore oh wait I need to buy more storage and migrate everything
I don't agree. There are a lot of services which provide fully managed platforms with multiple competent providers you can migrate between. WordPress + CPanel does a great job at this, I've seen barely technical folks manage multiple websites and blogs,export and restore backups, migrate between providers. And this was all 10years back.
FreeNAS supports docker, so arguably, clicking download on some panel somewhere to get a working locally hosted [wordpress, game server, ...] is simplified.
"simplified" things with Docker and k8s.. if you are using docker for anything other than microservices and development environments you are doing it wrong.
Containers are substantially different to VMs, but in practice I see them as equivalent technologically and operationally for my home lab/self-hosting use cases (sans weaker security guarantees, but that I can live with).
They are a simplification in practice, because I can toss everything into a bunch of Pis that stay on 24/7 rather than worry about the power use (and the power bills) that would come with using a more fully-featured rack server/old PC that has enough RAM to deal with multiple VMs.
Containerd then? Containers predate docker in the form of VMs, the only difference being optimizations with not having to run another entire OS kernel to get software running in the exact same state on many machines in just a few seconds.
You can always pay for managed versions of these services if you want the piece of mind. NextCloud and email providers come to mind.
Other than that, Docker has simplified things a lot in the self-hosting space. It's not quite as plug-and-play as Lego bricks yet, but it's made experimentation and sustainable self-hosting much easier as long as you're running on a supported platform. It's much easier to version the environment, much easier to keep the application runtime configuration clean, and much easier to move service instances across machines if you're too small to invest time in orchestration like K8s.
ahh what. You mean running Hugo deploy to an S3 bucket with a cloudfront distribution infront. It costs like 2 bucks a month.
If you're blogging your content is perfect for static site generators. And there's other options that can make it even more simpler than that.
Hugo, s3, cloudfront? That sounds neither simple, nor well suited, for a lot of use cases like the local bakery or artist that wants to maintain a site. Most of the static site generator ecosystem seems to be aimed at developers that want too dabble in solutions way oversized for the problem. I don't need global CDNs for a site that sees loads easily served by WordPress or some other non-sexy CMS on a single instance of a random HTTP server, I need a CMS that I can put on there in 5 minutes, create an account and let people administer content. Sure, I realize there's a) valid use cases and b) options for that in the SSG space but quite a lot of the development there seems like solutions for imaginary or self imposed problems the world at large has moved on from.
Try https://netlify.com. My mother runs a small business, I moved her website from a shitty HostGator setup to there and it's been a charm.
Also, I run my company's website on it (https://assembl.net) and my blog (https://blog.slm.space). Very different sites, but they all run through Netlify. I haven't had a single problem with them.
I think even the most tech-illiterate could use it with their drag+drop site hosting feature + https://forestry.io CMS.
Every time I break out of its ecosystem, I'm simply horrified. Using an external service from AWS or Google is a nightmare of configuration. They are both preparing on day 1 for your application to become Facebook and Amazon, forcing me to make decisions I don't even begin to judge, and opening pathways through security that might be right or might be complete disasters. The "happy path" for a new user is rarely clear, and the documentation is bewildering because it's aimed at much more difficult cases than mine.
Netlify does a great job of simplyfing to a small number of likely paths. At some point I'd surely break out of its use case and be forced to swim in the much deeper pool of a more general provider, but at least I'll have some grounding to start from.
Netlify is one of the few tools where everything is as I expect it. Everything I've done with it has 'just worked' and I love the automatic preview builds that are generated for GitHub PRs.
Static website generator with a markdown editor will end up cheaper probably.
Wordpress or drupal will generally end up easier as long as nothing goes wrong.
From my experience with inexperienced use and users stuff will go wrong at some point and then it can be a bit more of a hassle.
They are definitely cheaper, as long as you already employ technical staff.
I've worked as an IT volunteer in a small org, and there was no dedicated tech beyond me. Any kind of setup that I built for authoring the website would have to be as usable by non-technical users as possible, so I went with Wordpress. I had no replacement and there was no-one with enough technical know-how to handle authoring and troubleshooting a cheap static site, so I deliberately chose WP to minimise problems. The cost was 2-3x larger than the cheaper alternatives, but ultimately it was keeping the website up that mattered.
I agree that static site generators could really use some better quick start features. But i disagree that turning on WordPress is any less complex. There's a lot more moving parts that your responsible for with wordpress than having a folder full of markdown files. The benefit of a CDN is the fantasy it gives you of people reading your blog en mass. And also it's all managed elsewhere and they handle ssl for you easily.
That's fair, I guess I'm a bit biased there by the ecosystem I've been using for decades. I was just thinking of this old stack as a nice middle ground, it's so boring pretty much every hoster is good at managing the major moving parts without any lock in.
Or if you need a DB, then use SQLite. It's a file that's also a relational database, and it's better than any file scheme you can come up with for this purpose.
It is really not that hard... I spent less time setting some nginx / letsencrypt than trying to configure google loadbalancer for static files with https and their blob storage...
It likely will have a better uptime too and require less maintenance over the years
That is a nice quote, but not helpful. This is a scenario that can happen and does happen, and that needs to be taken into account. Running and supporting anything of appreciable complexity and keeping it secure and legal is not free. I self-host a bunch of things and I do feel this pain quite acutely at times. This can be a very strong argument in favor of SaaS snd should not just be brushed aside.
I don't think sites like blogspot and medium says it is their content. Also, on the flip side I know that most of the self hosted sites dies in 5 years. You can say that it is their fault, but statistically speaking I would pretty much assume that the site will die. Thirdly, self hosting is not that easy. Even most developers do not have complete knowledge of things like server management, security, migration, networking etc. let alone other users.
> You can then switch providers, switch systems you use, or whatever you want to do.
Switching from say blogspot to medium is much more easier than even switching between two regions of AWS. Switching cloud provider is tough.
> Even most developers do not have complete knowledge of things like server management, security, migration, networking etc. let alone other users.
You don't need to! Managed Hosting is cheap as it can get!
2 €/month will get you 25 GB of web storage + 25 GB of mail storage, a .de domain (in my case at least), SSL through "Let's Encrypt", PHP, MySQL, 100 mailboxes, full control over the DNS records and whatnot else.
For 3 €/month you get three times of that.
Hosting is really dirt cheap and competent people will take care of all the messy stuff you don't know about.
As written further down, I'm currently going with netcup.de, which some consider kind of a "discount hoster", but it works for me.
You can also choose hetzner.de, which I think is somewhat of a "higher level" hoster and it's still pretty cheap: https://www.hetzner.de/webhosting
2 or 5 €/month is just worth it for me. Even if I could get a VPS for free, patching it, making sure everything runs just fine would take more time than it costs me with any (German) hoster who takes care of all the things I don't know about.
Right now I'm going with netcup.de for normal hosting, but I got some vServer cloud whateverthingies running with hetzner.de (they also have managed hosting for < 2 €/month) for a PiHole and tinkering.
I've been a happy customer of all-inkl.com for like 10 years or so, mainly switched to netcup because they were cheaper.
But all-inkl is great for business, 24/7 phone support and you talk to REAL people and technicians call you back for no extra cost and things like that.
I'm working on a blogging platform which of course allows you to add your custom domain(s).
Another feature we have is being able to add any URL you like to your posts or even redirects to the "main URL". So for example if you migrate from WP and some post had this URL:
domain.com/posts/123
You could use that as your main URL, or use it with a redirect to a new URL such as:
domain.com/hello-world
We're also working on a feature to be able to download all your content in a .zip file with HTML files, markdown files, and images.
Are there other features you'd like to have to feel "in control"?
We will have themes with settings. We would like to support custom themes but certainly not at launch. It would probably make more sense to work as a headless CMS so you can plug it into your SSG of choice and have total control.
> We're also working on a feature to be able to download all your content in a .zip file with HTML files, markdown files, and images.
Can you go to another instance of the platform, upload your zip backup, and restore everything?
I've had that feature on my blog for years[0], and backup/restore feature has come in handy so many times, from setting up a dev environment to outright saving my bacon.
We're not WP or Ghost, so there is really no other instance of the platform. You could copy paste the files on the zip and use them with any SSG though.
We would also like to have a WP plugin so you can import the zip into a WP install, although there's so much to do right now that this is not really a priority.
Like you said, get your own domain. But crucially get a good unique pseudonym before you start. Don't get your own name, unless it is really unique.
Your pseudonym should meet 2 criteria:
1) There shouldn't already be any obvious search engine results associated with it
2) It should be unique and easy enough for the common person to recall (aim at least to be recalled when you see the name on a list of multiple names but ultimately to be recalled from fuzzy memory directly or by means of searching for a key word)
>If you post on someone else's domain, then it's their content, and they're in control of it, not you. No matter what is claimed otherwise.
Likewise, if you are paying for hosting and a domain to any corporation, they effectively control your published content. Isn't that so? A hosting company can remove your account; a domain registrar can ban you, too. Isn't that what had been happening to Gab?
The only way you can be in full control of your own content is publishing a hidden service on Tor, an eepsite on I2P, a site on ZeroNet--provided you use it with Tor--, etcetera. Sure, it's availability at the expense of visibility, but that's how it is right now.
Yes, but if you have the content and maybe even a static site generated by Hugo or something on a computer you control, it doesn't take more than few minutes to deploy the whole site to a new host and transfer the domain to point to the new address.
What if your content is deemed too controversial or "problematic"? Will any company host it? Any domain registrar? Many topics now considered unpublishable weren't, not that long ago. And that's a slippery slope I wouldn't dare to test. The free flow of information must prevail if we want to preserve our freedom.
I don't think anyone's capacity to publish anything online should rely on trust nor gatekeepers.
Given your initial example was Gab, I'd say that if we want to preserve freedom we must have the ability to make those things unpublishable.
To quote Karl Popper:
"Unlimited tolerance must lead to the disappearance of tolerance. If we extend unlimited tolerance even to those who are intolerant, if we are not prepared to defend a tolerant society against the onslaught of the intolerant, then the tolerant will be destroyed, and tolerance with them.—In this formulation, I do not imply, for instance, that we should always suppress the utterance of intolerant philosophies; as long as we can counter them by rational argument and keep them in check by public opinion, suppression would certainly be most unwise. But we should claim the right to suppress them if necessary even by force; for it may easily turn out that they are not prepared to meet us on the level of rational argument, but begin by denouncing all argument; they may forbid their followers to listen to rational argument, because it is deceptive, and teach them to answer arguments by the use of their fists or pistols. We should therefore claim, in the name of tolerance, the right not to tolerate the intolerant."
I don't think it was necessary to quote anyone in order to make such a simplistic point across, though I get why anyone would do that--to appeal to authority, of course.
In any case, and even though I'm not advocating here for "unlimited tolerance"--only freedom of speech, which is not one and the same--I'll bite:
>But we should claim the right to suppress them [the "intolerant"] if necessary even by force; for it may easily turn out that they are not prepared to meet us on the level of rational argument.
Who is "we" and "us", here? Who gets to decide which community or group, which kind of action or speech, do "we" have the right to supress, "even by force"? What criteria are "we" going to use to do so? Will it be chosen through a majority vote? Sociocratic consensus? A deliberative process, perhaps? Where do we put the limit on when to act? There may be groups of people or ideas that "we" could consider dangerous enough to supress them long in advance--you know, take some risks with potentially innocent people. Let's supress political dissidents, as well; "they" could be dangerous for "us".
Look, most of what is deemed "intolerant" or "offensive" isn't a clear-cut example of an existential threat to anyone or an evident danger one has to extirpate before it's too late. Moreover, the criteria is highly unstable, prone to rapidly change over media manipulation, corporations' interests, historical circumstances, culture, etcetera. I'd prefer letting everyone speak and act freely, acting with the full force of law when anyone's integrity is clearly and directly at risk, and never before.
Consider hosting in a country less politicized on the said issues. I think for example you could host most traditional views without the any issue in Japan, for three reasons: freedeom of speech is guaranteed by constitution, they likely won’t bother with foreign language and the said views are mainstream here.
Given how the problem of free expression is becoming in the West, there is a business opportunity on medium-term I think.
If you post on someone else's domain, then it's their content, and they're in control of it, not you. No matter what is claimed otherwise.
Yes, it's harder. Getting visibility is also harder. It's still the only solution if you want to be in control.
It's okay to pay someone to run the blogging platform, etc., as long as you own the domain name & can decide what it points to. You can then switch providers, switch systems you use, or whatever you want to do. But if you don't control the domain name, someone else is in control.