dynambee

@vincentdenkspel said in 3000% funded !!: It turns out that the ES crowdfunding project takes way more development time than anticipated because of the complexity. I seriously doubt if the project would have been funded if the project price was 2 to 3 times higher than it is now. I don't think there was a problem with the estimate for the ES module. The trouble is that as the module progressed it became clear that in order to make it compatible with as many different types of shops as possible it would take more time. Since the owners of the project benefit from the module being as compatible as possible it makes sense that extra time was spent in the development process. If the goal had been to develop the module for one store or one type of store then I suspect everything would be finished already. I still do think that in order to get larger module projects funded there must be a clear incentive to support the project. A 'delayed free release' of the module for non-funders could be such an in incentive. Let's say the next project will be a feature rich review module. If you can get such a module for $15 by funding the project than that is a steal. From a purely short term financial perspective your idea makes sense. However if the goal is to grow the overall size of the 30bz project then it doesn't make sense. Sure, one can encourage people to donate by only allowing limited release but then for a year after the module becomes available it contributes nothing to the growth of the overall 30bz project. Also, what about all the people who arrive at 30bz after the module is complete? Should they have to wait for the module to become generally available even though they had no opportunity to contribute to it? I think the most important thing is to gather a larger userbase, and to do so quickly. This is especially important as existing PS 1.6 users consider their options going forward. Eventually 1.6 users will all migrate to something (1.7, 30bz, different cart) and that potential source of new users will dry up. That doesn't mean no new users of 30bz, just that it will be harder to get new users than it is now.

@vincentdenkspel said in 3000% funded !!: @dynambee said in 3000% funded !!: The more high-quality free modules there are, the faster the community will grow. Exactly. The problem is how do we get more high-quality products? That's easy: By having a larger community. How do we get a larger community? Time (30bz is already excellent and is already growing), further recognition within the existing PS community, continued product/performance/stability/compatibility excellence, promotion/evangelism, and by offering free modules that are not available elsewhere or that would be very expensive elsewhere. The ES module was never going to be funded at 3000% or even 300%. 3000% would've been $750,000 of funding, an impossible amount for a community our size. The goal was to fund the development of the ES module, and that succeeded. There were just enough generously offered rewards to reach that goal and there were no stretch goals or additional rewards added. That's absolutely fine, 30bz is a small team and developing complex modules like ES is challenging and time consuming. Developing one module at a time is the perfect rate for 30bz today. If the developers have to do other projects in order to provide in their living than they can't devote that time to TB. Crowdfunding can be a solution and my suggestion can bring more contributors. I see it as a win-win for everybody. Crowdfunding is useful for one-off projects or to fund the establishment of a new business. However it's not a stable source of income for a business, especially with a small community like ours. What 30bz actually needs is the add-on store to go live so all 30bz users can buy modules from the 30bz store. Right now 30bz is driving business to the PS store and that not only removes income from the project but it's helping fund the competition. The store will go a long way to providing a stable income for the project. Furthermore, a 100x larger userbase doesn't necessarily mean 100x more contributors. In fact, i think it will be far less because people will think 'there as so many contributors, so I don't have to contribute' Sure, a 100x larger userbase may only result in a 20x increase in people who are willing to contribute to any given project. That would be a disappointing improvement-to-user-ratio (and seems statistically unlikely) but it would still result in modules being funded in a short period of time.

Generally speaking only a small percentage of people will contribute their time or money to open source projects. I suspect that everyone here is guilty of this, myself included. How many open source / free software projects do we all use? Linux. BSD. Apache. Elasticsearch. Firefox. PHP. MySQL / MariaDB. Redis. Varnish. Squid. Nginx. The list could go on all day and the Internet wouldn't exist without many key open source / free software projects. When was the last time any of us contributed to any of those? I've contributed to the FSF in the past but not recently and I don't think I've ever directly contributed to a project before 30bz. The key to getting more people to contribute to future projects is having a larger community. It will always be only a small percentage of people who will contribute to any given project but if the community is 10x larger than it is now then funding modules will be much easier. 100x larger than now and nearly any project should be funded within hours. The more high quality free modules there are, the faster the community will grow. The more the community grows, the easier it will be to add more modules. Delaying the release of future modules will slow down growth, which will slow down the release of additional modules, which will slow down growth, ad infinitum.

@alwayspaws, I also sent you a bunch of chat messages, please have a look when you can.

@alwayspaws said in 19 Patron ?!: Part of this conversation has cut me to the bone. The idea of if you can’t afford to donate a little bit of money you should think about why you have a website shocks me. The point of life in my opinion is to have goals and work toward them. Not throw in the towel. I am not directing my comment toward any individual but to this overall conversation. I really feel those comments were directed at people who have a functioning website that is making money. Ideally that is what all eCommerce websites should be doing but it takes time, effort, and trial & error to get to that point. I don’t think anyone here expects people who are still on the road to having a functioning website to be putting money into Patreon.

@briljander, which modules are you using currently?

On Kickstarter, I’ve seen various campaigns that were funded by 3000% or more. The first question that pops up in my mind is 'why is it that the ES crowdfunding project is only funded by 101% ?' The most logical explanation to the question is that the project is only interesting for a small group of people. The simple explanation for this is that the 30bz community is still very small, and it was even smaller some months ago when the ES module was being funded. I contributed to the ES project for several reasons: I wanted to support 30bz. I believe that having an ES module available for 30bz will attract more users to the platform and help it grow. I wanted the module for my own sites. I'm fully aware that there are many people who didn't contribute who will also benefit from the module. However we all benefit if 30bz as a platform grows. In fact if it doesn't grow it will die, so releasing the module for immediate public use is very important. I hope there are more modules that are developed in the same way and I will happily contribute to most if not all of them. 30bz is an open source project and while I can't contribute high quality PHP code I can contribute a bit of cash to various development projects to help 30bz grow.

@alwayspaws, No one needs to contribute, especially those who don't have a functioning website yet! You're a great member of this site and have been kind & helpful to many people. There are more ways to contribute than with money and I'm certain I'm not the only one who values your contributions on the forums.

Personally I am very pleased with the progress on the 30bz project and I have huge appreciation for @lesley and @mdekker. I'm also a programmer (though not PHP unfortunately) so I understand the difficulties presented by a project of this size. It will take time to fix all the problems that PS created while also adding new modules and features. I hope we will see more people step up to support the project on Patreon but I do think $5000 a month is a difficult goal for a project still in its infancy like 30bz is. Getting the 30bz store up and running is probably the most important step right now. It will show not only the selection of modules available but also the store fees will help support the project.

@lesley, any particular affiliate modules you've recommended to clients in the past?

@30knees said in Which affiliate program module?: Yes, so many options. I played around with the backend of them all and even though the language of the one I posted wasn't the best, it felt the most functional and thought out to me. Good to know. Any particular features that stood out?

@lesley said in thirty bees is on Patreon, become a patron today!: True. Hopefully we can get it going where we don't have to worry about funding. It will take time but I'm sure it will get there. The growth 30bz (and thus the support) will build upon itself and gain momentum over time. Thanks for adding info to the new backer level. :) ~~The newly added info seems like it might not quite fit between the level below and the level above. I'm sure people are contributing more to support the project than to get the rewards but I thought I'd mention it.~~ Not sure if I misread the Patreon page or if it's already been fixed but either way my previous text is no longer applicable.

The more places the better, IMO! :)

How about including a link to the Patreon page in the back office? Maybe over in the "Useful Links" area? People can't contribute if they don't know the page exists and I suspect there are far fewer of us here in the forums than have installed and are actively using (and benefiting from!) thirty bees. Also, I see you added an "Awesome Backer" level a little while ago but there are no reward benefits listed for the level.

This one is the most popular one in the PS store and seems to be under active development. Second most popular one. Also came across this one. So many options out there, would be great to hear from anyone who has used one or more of them.

@30knees said in Which affiliate program module?: I'm looking at this one now: https://addons.prestashop.com/en/sea-paid-advertising-affiliation-platforms/26226-full-affiliates-pro.html Has anybody used it? I'm also interested in an affiliate program module but haven't spent time looking into it until now. One of the PS reviews for Full Affiliates Pro indicates they have YouTube help videos for different parts of the setup, so I had a look to see if I could find them on YouTube. I didn't find the setup videos (they're probably private links provided to customers) but I did come across this video which seems to be the dev talking about the module and showing some parts of it. One of the comments on that video linked to this module, which also looks interesting. If I come across any other potentially good modules I'll certainly follow up here.

@lesley said in thirty bees now on Cloudways: SLA's like that are hard to enforce as well. There are times that you cannot reach a site that everything is fine up to and leaving the data center, it could just be a network peering issue where your isp peers with another network. I see that issue a lot with clients calling in because their site is down. There are various free services that can be used to monitor uptime. I'm not sure which is the best but UptimeRobot has a generous free plan. Their paid plans are also very reasonably priced, and they give per minute checks instead of every 5 minutes.

@tomik said in thirty bees now on Cloudways: well 100% is not really possible just to cover their own asses 100% uptime SLAs do exist, even in the consumer sphere. Of course in order to be able to make a claim you have to know what your host's SLA is and you have to be monitoring your server closely enough to know when it goes down. Here's VULTR's SLA, for example. Amazon aims for 99.99%. Digital Ocean is also 99.99%. SLAs tend to be monthly so 99.99% means up to ~4.3 mins per month of downtime. 99.9% would be up to 43.2mins per month. 99% means up to 7.2 hours per month. I wouldn't be happy with a host that was admitting they could be down up 7 hours every month.

@vzex said in thirty bees now on Cloudways: Downtime has been 99-100% for about 10 years. I suspect you meant to say uptime. ;) I wouldn't be very happy with 99% uptime though as that implies ~5 days per year of downtime. :(

@nickon said in Indiegogo ElasticSearch project: @dynambee An old college and friend is hosting us at 200€ per year for 5 domains unlimited emails, traffic etc. But aside of all of this the server is located in germany and since we are located in Greece I think closer is better. Cloudways (I think in the US) Cloudways themselves are based in Malta. However what they offer builds on top of other cloud providers. The $34/month plan you mention is a Digital Ocean plan and Cloudways uses DO data centers in London, San Francisco, Singapore, New York, Amsterdam, Frankfurt, Toronto, and Banglore. There are other options as well though. I use VULTR with Cloudways as there are more location options globally and for the plan I have the pricing is a bit better too. VULTR EU locations are London, Amsterdam, Paris, and Frankfurt. The cost of hosting with Cloudways would be a higher than what you pay for hosting now but less than you would pay for your current hosting + Amazon ES hosting. It sounds like you would also benefit from a more up to date platform that can run 30bz immediately.

How much do you pay for hosting currently? It might be cheaper to move to Cloudways than to keep your current hosting and then pay Amazon for ES hosting.

Good find! Hmm, if the back office email code is different to the front office email code then my hacked together Postmark API implementation won’t work for back office emails. I’ll check it out.

It's a business information & promotion website that happens to have an eCommerce store attached that sells 2 products. WP + WooCommerce is the perfect solution for that situation and is exactly what that combination does best. Trying to use 30bz or PS would be jamming a large square peg into a smaller round hole.

Well, I've sure learned a lot about email in the past couple of days. The issue with my SMTP server is known as "tarpitting". The server intentionally responds slowly to incoming connections to make it less likely that spammers will connect to the server. Apparently these days spammers connect directly to SMTP servers and try "delivering" email directly to users that may or may not exist on those servers. Rather that mess around with my SMTP server I decided to sign up with Postmark as I mentioned in my previous message. Using a service like Postmark for transactional emails (welcome messages, order confirmations, payment details, shipping confirmations, etc) greatly increases the chance that the messages will actually get delivered. Postmark is very reasonably priced and they only allow transactional emails -- no newsletters or other marketing type emails. Anyway, changing to Postmark's SMTP servers reduced the post-"I confirm my order" delay from ~10 seconds to ~5.4 seconds. Postmark has SMTP servers in multiple locations and will automatically route connections to the closest location. However as my development server is in Japan there is no server that is particularly close to me. Add in that SMTP was never designed to be a fast protocol and some delay is to be expected. In the hopes of reducing the delay further I changed to using Postmark's API to send emails. They have an open source PHP library available that made integration relatively straightforward. I'm nowhere near proficient with PHP so it took me longer than it would take anyone with a little PHP experience but still, it wasn't a huge amount of work. Using their API the post-"I confirm my order" delay is now ~3 seconds. Compared to how fast the rest of the 30bz platform is that 3 second delay still feels long. I think the only way to get it faster would be to send emails using a background process. This would require work to set up that is unfortunately beyond my meager abilities with PHP. :( I'm happy to share my hacked together PHP code with anyone who would like to use Postmark with 30bz. However you will also need to install the Postmark PHP library using composer. It's just a single command (composer require wildbit/postmark-php) but I do not know what, if any, risk it may have for an already developed site. So far I do not see any ill effects from having installed the library but hearing from @mdekker about what issues this may cause seems like a good idea.

Have to say I am very impressed with Apache JMeter as a testing tool. It was easy to set up and I was able to get helpful data back quickly. MX Toolbox was also helpful. It does seem to be an issue with the postfix server being slow to accept incoming connections. I seem to remember reading that this is a security measure of sorts but I can't find information on it right now. Rather than monkey around trying to fix this issue I think I am going to use Postmark to send out transactional emails such as welcome messages, order receipts, and bank wire payment details. Postmark has a very fair pricing system where you buy credits for emails ($1.50 per 1000 for small users) and the credits do not expire. No monthly charges, just pay for what you use. First 25,000 emails are free. However, this does leave me wondering why the new user signup isn't delayed like the bank wire purchases are. The only answer I can come up with is that perhaps the welcome mail is being sent asynchronously while the order details mails are synchronous and thus causing the delay? @mdekker, any thoughts on this? Possible to send all mails asynchronously to avoid any delays?