Forum Activity for @dannya

alt=
@dannya
11/24/15 07:08:26AM
584 posts

Replacing image uploader


Jamroom Developers

Ok. Thanks. This is a bigger problem for other places where the core image uploaded is used though. We'll figure it out I guess.
alt=
@dannya
11/23/15 10:38:22AM
584 posts

Replacing image uploader


Jamroom Developers

We have implemented a custome image uploader tool with our custom modules. However, I want to use it with the jrblog module as well. However, the module uses the jrcore files. How can I replace the image uploader with my own code without using the form designer? I want to make sure it doesn't break when I update jrblog.
updated by @dannya: 02/24/16 04:27:24PM
alt=
@dannya
11/18/15 08:19:12AM
584 posts

Marketplace updates


Installation and Configuration

Like I said, I put support aside. And I'm happy to pay for support. That too is not based on how many servers I have. For the most part, I think Jamroom is an great value. I think I've shown my support by buying almost every release and module over the last 10 years; since version 3. The pricing for a single server is very reasonable and you guys do a good job supporting people despite a small team. However, I have also spent thousands for developer time trying to figure out stuff because of lack of documentation. Granted, single server install is your target market.

I'm just telling you, as a customer, and as the former product manager of the largest open source video platform on the market, the pricing of the cloud modules is not affordable for its intended use. Cloud services are designed to be deployed on many many servers. Your cost per server to run a microservice does not match the development cost. Like I said, the cost to develop it is the same weather I run it on 1 or 1000 servers. I personally think, the server cloud apps should cost money, and the clients/workers should be free. This allows people to scale at a reasonable cost. Again, think of the use case. If I need to deploy 1000 conversion workers, how much would this cost?

Again, just my opinion. But I can tell you I'd sooner develop my own skin than pay for $29 to put a skin on each conversion worker. I can't afford that. Appreciate the debate though. At least you care enough to do so.
alt=
@dannya
11/18/15 07:53:20AM
584 posts

queue depth


Using Jamroom

Ah. Ok. So stupid of me.

One more question. Is there a way to add data from a different server to your dashboard? Specifically, I want the queue depth from my queue server and the five minute load from my conversion workers to display on my main dashboard.
updated by @dannya: 11/18/15 07:53:34AM
alt=
@dannya
11/17/15 06:13:34PM
584 posts

queue depth


Using Jamroom

I'm sorry, but I can't find the queue viewer on the main server. Where do I find it. Sorry for the stupid question.
alt=
@dannya
11/17/15 05:51:22PM
584 posts

Marketplace updates


Installation and Configuration

Because the cost of code does not increase as my usage scales. It's a one time sunk cost.

Amazon has recurring cost every month because you are
1. Paying for facilities
2. amortizing cost of hardware
3. Bandwidth (and the infrastructure to support it)
4. People to administer it
Basically, there is an ongoing cost to them an the more you use, the more it costs them as well.

Software on the other hand is a once time cost (putting aside support and future development).

Lets say it to 10 hours to develop that skin; its pretty bare bones. Once you wrote it, it's done. If the time you invested was worth, say, $400, then you need to sell 14 copies to before you start making a profit. Every copy after that is pure profit. You can sell 1000 copies and its still only 10 hours you put it. Pretty good product.

On the other hand, look at it from my side. If I'm going to deploy 1 cloud server, $29 is not a big deal. If I need to put this on every cloud server its going to cost me more and more. Any thing over 13 servers, It would make more sense to develop it myself.

In a service oriented cloud architecture, where you distribute tasks across many many servers it is not unlikely that you could deploy thousands of servers, even for short periods of time. This is the value of cloud and what enables you to scale. However, you are adding a cost that makes it uneconomical to scale to a product that is intended to help you scale.

Another way to put it is that you are taking something that had a fixed cost to produce but increases in cost as you use it. In that case it's better to write my own skin or risk paying thousands of dollars for of something that has a one time fixed cost.

I don't purport things should be free, just reasonably priced. This pricing model is not appropriate for its intended use. It does not cost you any more time or energy if I deploy the skin on one server or 10000 servers.

Hope that helps.
alt=
@dannya
11/17/15 04:51:00PM
584 posts

Marketplace updates


Installation and Configuration

I think its fine to pay for it once as part of the bundle. It just the other servers in the cloud don't need the whole bundle. Usually it's just one or two cloud modules. My immediate concern is the conversion workers. Typically, a encoding/transcoding worker should be very lean. A quad core server is only processing a small number of jobs simultaneously. If you are processing large files, this means you will need to have many servers ready to go to handle spikes. To build a cluster of conversion workers, not only do you need a full jamroom install, but also multiple cloud modules. You're looking at $100+ to run ffmpeg. Adding another $29 for each just add insult to injury.

I know you'll suggest getting the monthly "unlimited license" package. But by pricing it so high, that's really the only option. The clients really need to be much much cheaper.

As for the skin confusion, you could just make it clear that it is intended for cloud servers and not the main app. They'll find out very quickly that its not.

Just my $.02
alt=
@dannya
11/17/15 04:38:08PM
584 posts

queue depth


Using Jamroom

So the dashboard is won't look at the queue server, it's only the local queue

So is there any way to check the local queue if you are using the queue client to send some queues to another server? I can't tell what which jobs might still be pending locally.

Haven't been able to check out the beta. Trying to get 5.2.41 into production. Will probably wait til its out of beta.
alt=
@dannya
11/17/15 04:09:03PM
584 posts

queue depth


Using Jamroom

I'm not sure if this is a bug or not. My main server dashboard shows queue depth of 20. My queue server is a separate server and when I look at the queue viewer, its empty.

The only thing I can think of are queues that are not sent to the queue server. I'm not sure how to check those though.
updated by @dannya: 02/21/16 06:21:15PM
alt=
@dannya
11/17/15 03:54:23PM
584 posts

Marketplace updates


Installation and Configuration

Thanks. Not sure why it popped up now. It's been on there for a while and never asked me before.

I'm not sure how much it costs, but it should probably be free. We already have to pay for the respective cloud module. The skin just help keep the interface clean. e.g. we already have to pay for the conversion worker running on another server, it's tough to jusify doubling the price on each server just to have a clean landing page.
  15