Is "self hosted" considered a positive or negative these days? It seems like there is a small niche that still wants control over their data but everyone else seems ok with the cloud.
While some people will seemingly climb over their own mother in their rush to hand over their personal and business data to every man and his dog, some of us still have our sanity.
Having it available as self hosted doesn't exclude the possibility that the author or another company will offer it as a service. The reverse is often not true.
If I were developing a product like this, I'd build it for self hosting and charge a nominal cost plus yearly a maintenance/support subscription. Then you could deploy the same product yourself and offer it "as a service" for a monthly subscription. Atlassian does this for many of their products.
Also as for the business model you've mentioned. I've seen this done in competitive spaces like affiliate marketing. If you don't know most affiliates are super paranoid about thier data as leaking info about a campaign can lead to tons of copies and there goes your revenue. They seem to mostly prefer self hosted apps. Imobitrax offers a self hosted paid tracking system with a monthly fee. A lot of people used to use the free prosper202 self hosted tracking system too. Now recently I'm seeing a shift where people are not afraid to use cloud systems like voluum. I'm not sure what is causing the acceptance to change. It could just be that everything is cloud hosted now so people are getting used to the idea.
The problem I see with paid self hosted PHP apps, in this case, is that they are often using ioncube obfuscation and then if I can't read the source how do I know my data is more secure than a cloud provider?
I skipped over the free part too. Makes sense to be free and self hosted but this would also be cool as a "deploy to heroku" (or whatever cloud PaaS provider officially supports php). I still wonder if the creator has plans to monetize it?
My market research suggests that the market for hosted products is much larger but the market for self-hosted products these days is underserved.
There are people who for various reasons refuse to use any hosted product (legal issue, boss said no, etc.) and depending on the competitive landscape they may not have any other options than buying your product. This produces a captive audience.
I built a similar project management app (http://duetapp.com) and people love the fact that it's self hosted. This segment of the market is definitely smaller than the segment that prefers SaaS, but it's bigger than you might imagine.
Hey! I have always one doubt about this type of product (pay once). How do you distribute your source font, as well the installer to your customer? Do you use another software for a restricted area to download? Or is the download open to anyone, but you have to offer a license key to run the app?
The app is distributed as a single zip, but the installer requires a license key. Updates are distributed through the duetapp website, which also requires a license key before the update is delivered.
Nice! But is it secure? As the customer already have the source code, what stops him to change the license code? I mean, if the source code is available, I can install it by myself without using the installer.
It's definitely a positive, especially for enterprise. The cloud for enterprise means your' depending on people who dont know to do a good job, and it means you spend 3x as much time communicating vs getting work done.
I don't know. I work for an enterprise company and we're coming around to the idea of putting things in the cloud. I actually think the people running cloud hosted software are more of an expert in the software and devops than the people inside your organization in a lot of cases. Plus large enterprises have so much bloat that it can be way more expensive and time consuming to launch something internally. The biggest issue seems to be that these companies are still afraid of who controls their data.
Sometimes. All I can speak about is my personal experiences with a handful of enterprise customers and the times we moved things to the cloud. Communication is always an issue, we find that instead of spending time in house developing, deploying, administering, etc... we spend that same amount of time either one the phone or writing emails and documentation on what we need or what is wrong or what we need done.
We also find that the what the cloud team finds as a sufficient level of quality or service often does not match what we require or expect. And fixing that is a major PITA because it comes down to culture, and suddenly you are not dealing with your company's culture but a 'cloud' culture.
And on top of all that there is the issue of who controls your data. Is it secure? is it safe? What's your uptime, downtime, maintenance? Often we get promised good things (sales people are good at what they do), but it rarely works out as such.
The answer to this is find another cloud provider, but I dont want to spend months RFI, RFP, migration, integration, etc....again because the first cloud people were terrible. It's often just better in the long term to do it yourself.
It's a pay me now or pay me later scenario and I'll live by the words if you want something done right, do it yourself, especially for enterprise.
>these companies are still afraid of who controls their data
You make this sound somehow irrational.
For a large number of companies, business data is the key to the running of the business. Why would you ever let someone else control this data, much less someone who themselves doesn't have full control?
I wasn't trying to say it's irrational. I realize why companies would be afraid of putting proprietary data into someone else's hands. I still think the idea that it's going to be stolen or leaked from a cloud provide is not likely to happen. I'd be willing bet most of the time IP is stolen it's someone inside the company doing it. So much easier for a disgruntle worker to do it.
Do you not remember what happened with Codespaces?
How many times are there posts to HN: "AWS is down in <FOO>" with dozens of comments "yep, XYZ doesn't work right now"
How many service based companies have been bought out by a bigger fish only for the service to be shut down or changed in some dramatic way?
For the majority of businesses I very much doubt that theft is the primary concern with using SaaS. Control, visibility, access to your own data regardless of the app/system author's current situation, etc.
Even if theft is your primary concern - and I will grant you that for most small companies where it's targeted theft, the finger more likely points to an insider - using SaaS, especially if it's one builds on a third party provider's stack, means you are increasing the potential for breaches, and you don't even know what that increased potential is, and potentially you may not even know if a breach occurred.
Imagine a company that runs a service on Heroku (who in turn have built on AWS). So by using this service, you suddenly have great potential for individuals at three other corporations to heavily influence your ability to do business and remain profitable, either through data loss, downtime, or the aforementioned security breach.
I haven't even mentioned the issue of data privacy here - it seems quite common for modern SaaS startups to rely on VC funding to offset the costs of providing their service, with the vague notion of "if you build it, they will come... and then we can somehow turn them/the information we hold on them into a profit"
The terminology is perhaps a little messy, but I think "self hosted" and "in the cloud" are not necessarily mutually exclusive. On one hand, "in the cloud" can mean a SaaS platform where your data is locked in and you have no control over the application. However, I could also own the code and data and host an application myself "in the cloud."
This is the ideal compromise to me. I can own my data and the application, but I don't necessarily have to deal with all the annoying details of managing my own servers.
Yeah when I said "cloud" I really meant SaaS. You can certainly deploy a self hosted app to a "cloud" PaaS and get a greater ability to control your data than SaaS while still getting many of the other benefits a SaaS service offers.
Having it available as self hosted doesn't exclude the possibility that the author or another company will offer it as a service. The reverse is often not true.