jollyfrog

September 24, 2021

It appears, given the lack of response to the community invite to form a group, that there is no "community" interested in the theme.

Unfortunately, it does not look possible to design a theme based on the suggestions of the community when the community will not come together to vote on ideas and settle on some fundamental goals of the theme.

With this in mind, to end the endless discussion. I am officially dropping out.

If in the future, a group forms who are willing to make timely suggestions, vote on them and support the development of a theme, drop me a message.

For now, all the best.

September 23, 2021

8 hours ago, wakabayashi said:

@jollyfrog I am always open for some interesting discussion about ecommerce/tb. But I won't get much involved in any decision and surely not work. As I really plan do to my own thing. I am also not totally sure, what the plan is right now. A new theme the way you plan it, means basically, that you just don't like Niara Design. Is this correct?

To be honest, it wasn't that I had a dislike of anything. I explored TB on behalf of a client and realised that there wasn't a suitable theme available for their shop. I liked the idea of TB, regardless, and I like to contribute to decent open source projects if I can. I saw that certain members of the community were unhappy with the theme options available, and I have built themes in the past for various platforms.

I am no expert with TB, which is why I requested the help of those who have the skills and experience to implement the theme, but perhaps don't have the design skills to do it alone.

There has been a ton of discussion over the past 7 months and I took a back seat for a while, as there was too much confusion over what should or shouldn't be done regarding a theme.

The reason I am still in the picture is this thread keeps being resurrected, and I am repeatedly asked if I am going to design something. I feel that if the right people are willing to support, I should complete what I started/offered.

With that said, I have no personal agenda. I don't personally need a theme.

As a merchant and a developer, you clearly have an in-depth understanding of the problems which other merchants face (better than I do). I know you were looking at building something with Tailwind and potentially starting from scratch with your own theme.

If you feel that a new theme [design] is not the right solution, what would you suggest for other merchants?

Would you suggest that Niara should be improved instead? Or do you feel that Niara is sufficient?

September 22, 2021

1 hour ago, led24ee said:

If possible then add me too to this list.

I have created a list of group members and added you to the list. Thank you.

September 22, 2021

@yaniv14 OK. Understood

@cienislaw given the discussion over the last 7 months, I am inclined to agree with you.

With that in mind, I think it is best to ask:

Who would like to be involved in decisions/discussion relating to the new theme?
(anyone is welcome — no coding or design skills required)

In this group (looking at those who have commented on this thread and discussed the topic with me), I assume the following people would like to contribute suggestions?

@veganline, @Mark, @Traumflug, @datakick, @cienislaw, @zen, @wakabayashi, @DRMasterChief, @yaniv14, @Wartin, @AndyC, @toplakd, @superbiche, @Briljander, @Mediacom87, @netamismb, @Billy

Please make me aware if this is correct, or if you would like adding to or removing from the list.

Once we have a group decided, all (fundamental) suggestions should be submitted within a given time frame (prior to any designs or code being written).

I will do my best to document and summarise suggestions.

I assume everyone is happy that any disagreements can then be settled by vote.

Once the community has spoken, I will do my best to design the theme based on the community's wishes.
(It is unlikely that the design will please everyone, but hopefully the majority will get what they need out of the theme).

Is anyone available to support @yaniv14 in the near future, should he need it?

September 22, 2021

9 hours ago, yaniv14 said:

I will do it

That is good news.

How do you wish to proceed?

Are you happy to move forward with Orchid (in the current style) as a basis for the overall design, or would you and/or those interested like to make suggestions regarding the design and function?

September 21, 2021

@yaniv14

Yes, I would if the community agreed that it was the best course of action. It is a substantial amount of work to waste if there isn't an organised team who want to implement it.

September 20, 2021

@Mark I wish I had more to offer. I could provide a design, but as for integrating it into thirtybees, I just don't have the skills.

By the sounds of it, @wakabayashi needs another PHP dev more than he needs a graphic designer.

If someone needs me, I still get the notifications.

August 7, 2021

Hi @Script-Fu,

I started looking into the design of a new theme for thirtybees several months ago. Attached is part of a concept design called 'Orchid' (named after a beautiful, colourful bee found in Mexico).

The difficulty lies in my reliance on others with the skill to implement the theme.

My understanding is that the core team is very busy and cannot spare the time to focus on a new theme right now, and while others were kindly willing to get involved, the project lacked a decision maker. Many people, many ideas, but who should decide which route to take.

If I had the skills and experience, I would gladly build the theme myself, but my coding ability doesn't stretch that far and my knowledge of thirtybees even less.

On top of this, I understand that it is the view of the core team that a free theme should not be overshadowing the work of those who (do and could) develop (paid) themes for thirtybees as this may be off-putting to them and may hurt thirtybees in the long run, which makes sense. Perhaps the community could work on a paid theme... I don't know...

I am currently working on other projects, but if, in the future, the right support is there, I'll gladly do my bit to help.

February 17, 2021

I should clarify, while I have not attempted to manipulate Apache server logs, NGINX provides the following useful information:

IP address (to identify the user)
Dates and times the server was accessed
The URLs of the resources which were accessed
Referring domains
Response codes
The user-agent (which provides clues as to who/how the site was accessed)

I am aware, it is possible for NGINX to be configured to provide more data than this.

While the raw data could appear near unreadable, the following data could be filtered:

Number of visits over a given period
Most popular pages/resources
Average time on page
Common entry and exit pages
Whether the pages were accessed via other (internal) pages or accessed directly
New users vs return users (both as a percentage and cumulative data) over a given time period
Most popular pages by referring domain
Overall traffic by referring domain
...

To measure the data above, sensible assumptions must be made, such as "a session has ended once the user has not interacted for X mins/hours", and "the last page of the user journey does not count toward average time on page for any given page".

Admittedly, the server logs do not track button clicks or measure marketing campaigns or goals, for this you would have to make assumptions based on continuous testing and correlative data (changes to sales of a given product prior to, during and after a campaign).

Sales funnels would not be tracked automatically, but it is possible to track the journey of your visitors through the website.

There would be no keyword data relating to search engines, but Google has moved away from keywords and now focuses on topics (grouped keywords) and user behaviour (you can rank No.1 for a phrase without mentioning it on your website).

I agree that GA and Matamo provide more information and the information is more accessible. It was, therefore, irresponsible of me not to mention that there are other options. There is also Fathom, which I have tried, but the free (open source) version does not provide much depth of information, but could also be used for a quick overview/summary.

To sum up: Measuring server logs manually is not going to be feasible for everyone and, while not everyone want's to use Google, Matamo may be a happy medium.

February 17, 2021

1 hour ago, Wartin said:

Thanks for the data, Jollyfrog.

I use Matomo as I really don't like sending my customer's data to Google. It's not needed, with Matomo served locally I can know what people are doing and can choose not to save specific things like their exact IPs, for example.

Using grep and awk is fine (like Neo reading the Matrix XD), of course, but when more data is needed, I think it's really important to know there are other things out of GA.

@Wartin Yes, you are right. People need to know the options and Matamo would certainly be useful for ecommerce sites. Unfortunatley, with clients, I almost always have to use GA and other insidious software.

Server logs provide a wealth of information (not what I am sure Matamo can provide, but insightful) and are underestimated. The only issue is that the data is unfiltered. "like Neo reading the Matrix XD", indeed.

Working with clients, I became increasingly frustrated at their obsessions with keyword stuffing, content rehashing and backlink purchasing, all based on their analytics. It became difficult to convince them to focus on long-term goals and good practices, such as a solid content strategy.

On a relatively low-profile website, once content has gained pace and relationships with influencers have been established, the right post can deliver 10,000 to 20,000 visitors to the website in a few hours. Analytics usually cause website owners to look internally (rather than externally), it can be tempting to repeat what has worked, instead of exploring what could work better.

Of course, you can see that my issues are mostly with how people interpret analytics rather than the analytics themselves.

Progress isn't made with change alone, but by building on what you already have. I would, if I were running an ecommerce shop, want to gather more meaningful data than what I could get through server logs alone. Hence, I would like to know more about your experiences with Matamo. Also, do ad blockers affect Matamo in the same way as GA?

February 17, 2021

@datakick Apologies if I have misled anyone, that was not my intention. You are right, this does have very limited application. For ecommerce, people may desire/benefit from more information.

I spend a lot of time dealing with other people's websites and it is too easy to neglect my own. Server logs provide me a quick overview of my websites' performance which can be useful to me.

----

The section in question has now been removed from the post.

February 17, 2021

Just now, movieseals said:

I am not an expert but I am comfortable with terminal, so no worries there. I always intuitively knew that the server logs were more accurate but they also track ALL traffic, bots, spambots, etc. so it is hard to figure out what is useful and what is not. Thank you for volunteering this. Very interesting and useful.

You are right. The way I deal with the bots is to look for patterns and eliminate the associated IP addressses. Bots visit pages/links which regular visitors don't. They also, often, visit pages more quickly and visit far more pages than the average user. Hidden links can be placed on the site as a trap, and timeframes per page and number of pages visited can be measured. I also add to the block/bot list manually if one slips through and to eliminate my own IPs.

These techniques aren't perfect, you may miss some (slow crawlers with ever changing ip addresses) and you may get some false positives, but it will likely be more accurate than GA in the long run as it won't be blocked by ad blockers.

I run a single command which downloads the logs, checks for bots and gives me a basic analysis of the data (sorted bar charts for individual page views, total hits vs unique visits and top refering domains).

February 17, 2021

1 hour ago, movieseals said:

I would like to take you up on that offer. I despise Google Analytics and I relish the irony and the gall of Google telling us that our site needs to be speedy to get ranking yet their own tools are some of the biggest culprits in slowing down things! They are not alone, of course, but their tools are not optimized and raise flags everywhere for their slowness, including in Google's own speed tools!!!

I remove Analytics eons ago but using the server data, which is often confusing, to achieve the same results, I am all for.

@movieseals In that case I will prepare something and share it soon.

I am used to reading nginx logs, but I am aware most of the thirty bees sites are probably using Apache. That said, I am confident that this concept can translate.

One question: Are you comfortable in the terminal? (I use tools like grep, sed and awk to achieve the result).

Reading server logs has been very useful to me. So far, I read them regulary, but intermittently; but when I have time, I will automate the process.

February 8, 2021

5 Tips for Better Search and Sales Conversions

Hopefully, some of you will find this useful. Not everything in this guide will be as easy to implement in thirty bees as others, but at the very least it should point those who are interested in the right direction. If anything is unclear please reach out.

If enough people want to know more about any of the areas I have discussed below, I may create a more detailed post covering the subject.

1. Lighten the load (page load that is)

Reduce external requests

Say no to web fonts

Most choose to avoid using system fonts and generic fonts, such as "sans-serif", for good reason — they can be unpredictable in headings (causing your layouts to fail) and make your site's design inconsistent across devices.

However, the solution I see most site owners using is unnecessary and problematic. Web fonts solve one problem poorly and cause other issues.

Not only does loading your fonts from another server cause slow rendering of the page content, your fonts may also load too late (the user sees your fallback font first). This contributes to a poor user experience.

Google uses web fonts as a fallback for their analytics software — while there are many people blocking Google Analytics with ad blockers, few are blocking Google Web fonts.

What's the alternative?

Want a lightning fast page load? Serve your own fonts. This gives you the best of both worlds. Simply convert your [suitable] fonts (check the licence) to woff/woff2 and serve them directly with CSS.

If anyone needs more information on this, let me know.

Forget Font Awesome

Font Awesome is convenient — to you, but not to your customers. Yes, that's another server to wait for before your page can render.

Did you know SVGs are often tiny (in file size), and with [inline] SVG sprites you can load all of your graphics without a single request, and they can be cloned/partially cloned across your page? Now that is more awesome than Font Awesome!

Better images guidelines

The number one performance issue I see, which covers online businesses of all sizes, is poor use of images.

Here are some best practices to solve slow loading pages and related poor conversions:

Serve your own images (don't rely on CDNs and avoid hotlinking images)
Standardise your image dimensions- Keep all images under 100kb
Use srcsets in your code (where possible) to tell the browser which image to serve to which device.
Your images should be high quality and consistent

There are a number of open source programs to help you resize and optimise your images. If you are comfortable with the command line, I strongly recommend using imagemagick.

If you want to know more about optimising images for the web and ways to automate the process, let me know.

2. Be minimal and functional

Your customers likely care more about function than style. People are busy — they want to find what they are looking for, buy and leave. The best user experience on an e-commerce site is a fast one.

For this reason:

Make your design predictable and practical
- Your logo links to the home page
- Your contact page is in the main menu and/or details in the footer
- Your Returns Policy, Privacy Policy, Ts & Cs should be in the footer
- Your login button should be top-left
- Your logo, tagline and branding should make it clear what you do

Create a sensible link architecture
- Link to your most popular products and pages in your footer
- Have two site maps. XML (for crawlers) and HTML (this one linked to in your footer)

3. Blend in — take advantage of familiarity

Nothing says you are at the top of your game more than looking like you are at the top of your game. Don't try to be different for different's sake. Let familiarity be your friend. If you are new to the industry, research the leading sites in your industry and blend in. You should still have features that identify your brand and make you memorable, but too much personal expression and you will appear like the amateur in the industry.

The world's leading websites don't have flashy designs; they are, for the most part, basic in style, predictable and easy to use.

Ignore this rule only if the competition is poor and can easily and predictably be beaten.

4. Be better

So how do you stand out? Focus on your offering!

Are you the cheapest?
Do you deliver faster?
Are you the ethical choice?
Perhaps you have products which others cannot source?

If you don't know why your customers shouldn't buy from Amazon, neither do they.

As a general rule, if you want to sell more products than your competitors, provide more product information. The more information, the more confident your customer will feel in both the product and your website.

Also, this content WILL help you to rank higher on search engines.

Create elaborate pages about the collections you sell

Search intent is important to understand. This is possibly one of the most under utilised tools.

People often look for information before they look to buy. That means your product page may not be what they are looking for. If you sell wood burning stoves, share information about the brands, which stoves in your store are most suited for small homes, which stoves are the most offer the most "bang for your buck". Display the appropriate products on the pages and link to your products, you are likely to qualify a sale which wasn't accessible to you before.

Speak to manufacturers to see if you can gather more info on your products than your competitors.

There are numerous creative techniques I have used to achieve high search rankings and boosted sales conversions. If you are interested to learn more, let me know.

5. Be useful

Start creating content

Create [or ask others to create] guides which may help your potential customers. Digital Ocean (not an affiliate) sell server space, but part of what pushed them ahead of their competition was their content. They pay people to create technical guides as a way of on-boarding new customers.

What content would be useful to your potential customers?

Add to the blog (at least) once per week

If a visitor reads your post and returns in two weeks to see what else you have written only to find the same set of posts, they likely won't return.

Attract qualified visitors by helping them and you will build a loyal customer base.

If you want any advice on building a successful content strategy, let me know.

6. BONUS! The #1 Factor for Ranking on Google:

~~ Direct website visits ~~

Are customers recommending your website?
Do your customers return?
Do customers search for your brand name?

As I'm sure you can see, the right offering and a winning user experience are vital!

Regards,

——— jollyfrog ———

Graphic Designer &
Digital/Content Strategist

February 7, 2021

15 hours ago, Briljander said:

I think you also should consider to make the theme heavy configurable like many of the paid successful themes like Transformer and Warehouse.

I know @zen started to make a module or something earlier. Maybe he can be a part of the collaboration.

I guess there is a necessary balance we must find between configurability and simplicity (for the merchant). For example, Panda (created by the same developers as Transformer) is a highly configurable theme, However, most of the shops I have seen using it are poorly constructed.

Wix is also very adaptable, but it is mis-sold to business owners. In the ads they claim it is very easy to build a beautiful website. Yes, for a graphic designer! Also, contrary to popular opinion, beauty is overrated!

I have worked in digital marketing for around 13 years (I manage a small agency) and one of the most common issues I have to solve is business owner [personal] expression. That is, altering things that didn't need altering. A theme needs to be robust, competitively designed and useful enough to the merchants so that too much tampering isn't tempting.

UX is complex and simple in different ways. The layout and function of an e-commerce website should be as predictable as the layout of a brick and mortar shop. This leads to predictable sales conversions.

Of course, with that said, I believe there should be some flexibility, as not all shops function in the same way; and what is predictable to one market may not make sense to another. For example, digital assets vs physical products.

This theme doesn't need to compete with Panda or Transformer. This theme needs to compete with corporate sites in both search and sales. For the thirty bees merchants, these are the competition.

February 6, 2021

2 hours ago, superbiche said:

I'd be happy to collaborate with you in order to build this new theme. I used PS for many years in the past, and was quietly waiting for the announcement about TB's future. Now that I'm (quite) reassured about it, I'm ready to spend some time on it, as I'll probably offer it to future clients as their e commerce platform.

Awesome!

2 hours ago, superbiche said:

how do you feel about a Tailwind CSS based theme?

I have some reservations, but definitely open to hear and understand the arguments for it.

2 hours ago, superbiche said:

The only condition for me to spend time on this, is that we pick a license that's really FOSS and we commit to making it 100% free. I wouldn't be against earning a few bucks, say by helping show owners to customize said theme, but the theme itself must be 100% free, no freemium BS.

I totally agree! The theme should be free (libre) and open source.

2 hours ago, superbiche said:

@jollyfrog could you PM me so we decide where and when we could have a first chat about this?

Will do!

February 5, 2021

1 hour ago, veganline said:

Seobilty tells me that several belts in the same category look like duplicate content to a search engine; it would rather index the lot in one go. I have trouble describing each belt to make it look different to a search engine.

@veganline I assume you own veganline.com?

I also assume, correct me please if I am wrong, that you are concerned that your pages are not all indexing due to duplicate content.

The following search will show you a lot of what is being indexed by Google relating to belts:
site:veganline.com intitle:belt

To avoid domination of stronger domains, search engines tend to limit standard searches to 2/3 listings for any given domain. Thus, if someone were to search vegan belts UK, it is likely that only a couple of your pages will show up (probably those which demonstrate more favourable user behaviour, such as 'time on page' and 'lower bounce rate').

You do appear to have some duplicate pages (on first glance). For example .../made-in-poland & .../29/made-in-poland seem very similar.

I don't know if that is helpful. Apologies, if I misunderstood your comment.

February 5, 2021

My thoughts, are to build from scratch but using the community theme for reference.

(Although I am very much open to the general perspective of those with experience working on developing the platform and those who have been using it).

Why?

Efficiency
Personally, I find it is quicker and more straight forward to follow a single coding practise or style throughout a project than to patch old code and designs.

Technology
The community theme appears to be using old CSS standards for layout (still using floats). Building these layouts is a lot easier now that CSS has developed. I expect that I will be able to help create a theme which is of a high technical standard, which is very light.

CSS Grid, Flexbox, inline SVG, CSS animations, CSS & SVG sprites, srcsets, localised web fonts etc.

Freedom
The old theme doesn't need to be scrapped. Maybe it can be patched up, but a new theme should, in my opinion, help to support those who the current theme just doesn't serve (that isn't to say it isn't useful for reference).

Maybe a new theme could give the platform some (more) wind in its sails. I know that @datakick and others are putting in a lot of work behind the scenes, and I think that the front-end of the project needs some attention to motivate the community and to encourage other developers to support the team.

The new theme doesn't have to be twenty themes in one, but I see no reason why it shouldn't be designed to a commercial standard, while still being in the hands of the tb community. I haven't been on here for long, but I love open source, I enjoy contributing, and I don't stop until a job is done.

@cienislaw I agree with @zen, if those who are interested in a new theme vote on what features they want to see, this would be a good start. Something you love in another commercial theme?... anything is possible! So I would love to see what features you have in mind.

One thing to keep in mind

While the perspective of each individual in the community is important, for the theme to be competitive (so tb sites can compete with those built on other platforms) there must be a consistent, professional style throughout. Many ideas can be good ideas but some ideas conflict. Aiming to serve everyone will likely serve no one. So there should be a clear objective, mock-ups, approval from those who have been working on this project and a single-minded collaborative approach.

----

I am nobody, so no one has to listen to me. But when I come over for dinner, I like to wash up. I use open source software, and I feel I should contribute to say thank you to those who give their time to serve others (including me). At the end of it, if I have a new theme to use with my clients or for a personal project... Awesome!

February 5, 2021

I am an experienced graphic designer and, if others will support, I will help design a new (free) theme for thirty bees.

I have recently been building and manipulating (Ghost) themes for clients, but a couple of weeks ago a client asked for help setting up an online shop.
While I like the general ethos behind the project, I can see that themes are limited.

The catch: I am not a PHP developer.

I can, however, help with the following:

Graphical Mock-ups
Vector logos, icons and illustrations
Animations CSS and SVG (SMIL)
HTML & CSS (I could build the pages static if someone was capable to help bring the theme to life).

Now, I understand that this means someone would have to collaborate with me on this to make this work with thirty bees, and given that I can see that the few developers who work on this project have little or no time to spare, I understand that this may not be possible.

Regardless, the offer is there.

February 2, 2021

20 hours ago, Wartin said:

Oh, too sad, you tried a lot of things... Maybe @datakick know what's happening here...

It's OK, I managed to solve this with another tutorial and used the information on GitHub's installation guide to figure out what new dependencies needed to be included. Thank you @Wartin for your help. :)

@datakick If it helps anyone document the process, I would be happy to give my step-by-step notes to the thirty bees team. I am aware that you have a million and one jobs to do.

Plus, I can't be 100% sure if I have missed anything vital, so I wouldn't want it to be public without someone who knows more than me looking over it.

February 2, 2021

I installed again and, on completion, I noticed when I clicked on the back-end button there was a link and a message that my admin area had a new url.

The previous time I deleted the install directory too early and didn't see the message.

This appears to be solved. I assume you just use the new link form this point on and that it was generated for security. It would be good to make that more clear during installation to avoid confusion.

Just a suggestion. Aside from that, it looks like a great platform with many features.

February 1, 2021

I eventually gave up on following the Digital Ocean tutorial and the GitHub installation guide. I'm just not clever enough or experienced enough to fill in the gaps.

February 1, 2021

Hello,

I installed thirty bees on a vps and all seemed well until I clicked on the manage your store button and rather than seeing the back end, I saw the 404 message: Page not found

Does this mean that the install failed or is the link wrong?

Any help would be appreciated.

Tutorial followed (but switched PHP and extensions to v7.3 and used the latest version of thirty bees):
https://websiteforstudents.com/install-thirty-bees-ecommerce-on-ubuntu-16-04-17-10-18-04-with-apache2-mariadb-and-php-7-1/

## Install details:

Ubuntu 18.04 LTS
PHP v7.3
Apache v2.4.29
MariaDB v15.1
thirtybees v1.1.0

## Server:

RAM: 1024MB
CPUs: 1
Disk space: 25gb SSD

Please let me know if any further information needed.

January 31, 2021

5 hours ago, Wartin said:

It's OK. I'll explain how to do it in case someone else reads the thread.

You have to create a new file called, for example, mysqltest.php with the code I pasted. The username and password has to be changed, of course. After saving it, you have to visit that file from the browser, like: yourdomainname.com/mysqltest.php

If does not connect, it will show the error.

That seems to have narrowed down the possible issue.

I successfully used the script but it returned the error: Connection failed: Access denied for user 'newuser'@'localhost'

Yet the exact same username and password is successfully logging me in to MySQL in the terminal via ssh.

I'm thinking there must be a communication issue between PHP and MySQL.

Perhaps I was supposed to install something and I missed it or used the wrong extension?

In regard to the PHP extensions do you know what this means: "mysql (PDO only)"? --- this is taken from the requirements on the thirty bees GitHub page.

January 31, 2021

1 hour ago, Wartin said:
If I were you I'd try connecting to the database directly from PHP, like:
 <?php
$servername = "localhost";
$username = "username";
$password = "password";

// Create connection
$conn = mysqli_connect($servername, $username, $password);

// Check connection
if (!$conn) {
  die("Connection failed: " . mysqli_connect_error());
}
echo "Connected successfully";
?> 
(from https://www.w3schools.com/php/php_mysql_connect.asp)

Gook luck!

I'm afraid I don't know PHP. It doesn't look like I am going to get this solved, but thanks for your help. Take care.

Profiles

Forums

Gallery

Downloads

Articles

Store

Blogs

Posts posted by jollyfrog