Posted on by — 6 Comments

Art, AI and AI Art – Part 1

Alien lizards roller skating in a Pride march with Pride flags

Introduction to Part 1

This post began as a limited review of what has become known as AI Art. In order to do that, I had to delve deeper into AI in general. Consequently, the post has grown significantly and is now in three parts. This post, Part 1, looks at AI in general. Part 2 will look at more specific issues raised by AI art, while Part 3 will look at AI art from the perspective of an artist. Part 4 will be a gallery of AI-generated images.

Links to each post will be here, once the series is complete.

What is AI?

It has felt almost impossible to avoid the topic of AI in recent weeks. Hardly a day passed, it seems, without some pronouncement, whether promoting the next big thing or predicting the end of the world. One thing seems clear though. Whatever concerns exist, the push to develop this thing called ‘AI’ is now probably unstoppable.

Looking deeper into the topic, I soon hit problems. AI seems to mean anything, depending on the motivation of the person using it. As a descriptive term, it is almost useless without a qualifier. This isn’t unusual in public discourse. I’m not going to consider all the bad-faith uses of the term, though. I will limit myself to trying to get some clarity.

What is now called AI is different to the former idea of ‘expert systems’. Instead of depending on humans trying to input rules to be followed, AI uses the raw data to find rules and patterns. This has only become possible with the advent of so-called ‘big data’. The term refers to data that is so large, fast or complex that it’s difficult or impossible to process using traditional methods. This is why research AIs are being used in, for example, drug research or cancer screening, using data from health systems like the NHS.

Weak and Strong AI

Essentially, there are two sorts of AI, ‘weak’ and ‘strong’. For a summary of the difference, I turned to AI. Here’s what the free version of ChatGPT said:

The terms “strong AI” and “weak AI” are used to describe different levels of artificial intelligence capabilities. They refer to the extent to which a machine or computer system can simulate human-like cognitive functions and intelligence.

Weak AI (Narrow AI)

Weak AI refers to artificial intelligence systems designed and trained for a specific or narrow range of tasks. These systems excel at performing well-defined tasks but lack general cognitive abilities and consciousness. Weak AI systems are prevalent in our daily lives. They are commonly found in applications like virtual personal assistants (e.g., Siri, Alexa), recommendation systems, image recognition, and natural language processing. They can appear intelligent within their specific domain, but do not possess real understanding or awareness.

Strong AI (General AI or AGI – Artificial General Intelligence)

Strong AI refers to artificial intelligence systems with the ability to understand, learn, and apply knowledge across a wide range of tasks and domains, just like a human being. A strong AI system would have cognitive abilities that match or exceed human intelligence. This includes not only performing specific tasks but also understanding context, learning from experience, and reasoning about various subjects. Strong AI would have a level of consciousness, self-awareness, and general problem-solving capabilities similar to humans.

Despite the claims of Google engineer, Blake Lemoine, it seems clear that we are nowhere near achieving Strong AI (AGI). All apps and programs currently in use, including ChatGPT, are examples of Weak AI. They are ‘designed and trained for a specific task or a narrow range of tasks.’ In other words, they are tools, just as much as the power looms of the late 18th century.

That does not mean that the introduction of even Weak AI is without risk. Like all disruptive technologies, it will create stress as older ways of working are replaced. Jobs will be lost, new jobs will be created, almost certainly not for the same people. If steam transformed the 19th century, and computers the 20th, Weak AI is likely to trigger a wave of change that will be an order of magnitude greater. We are beginning to see some of the potential in health care. It is already aiding in diagnostics, drug discovery, personalized treatment plans, and more.

Concerns raised in AI ‘training’

The ultimate goal in training an AI is to enable it to find patterns in data. Those patterns are then used to make accurate predictions or decisions on new, formerly unseen data. This process requires huge computational resources and careful design. The vast amounts of data involved, whether in training or in day-to-day use, raises significant ethical concerns.

Ethical concerns

Examples include:

  • Biases present in training data are likely to lead to biased outcomes. There is a need for transparency and accountability in training of AI systems, but the sheer volume of data involved makes this difficult. AI can help but…
  • Privacy concerns also arise in the use of training data. Improper handling of data or breaches in AI systems could lead to identity theft, unauthorized access to sensitive information, and other security risks. AI could also be used aggressively, to identify weak points in data protection, to break in and generally to create disruption.
  • Facial recognition systems could increase security by identifying potentially suspicious behaviour but at the risk of loss of privacy. Bias in the training data could lead to inequality of treatment in dealings with some ethnicities.
  • Adoption of AI in medical contexts potentially creates risk for patients. Avoiding risk requires rigorous testing, validation, and consideration of patient safety and privacy.
  • AI-generated content, such as art, music, and writing, raises questions about creativity and originality and challenges traditional notions of copyright or even ownership of the content.
  • Self-driving vehicles may enhance convenience and safety, but ethical issues arise in, for example, resolving safety issues affecting the vehicle occupant vs pedestrians or other road users.
  • AI can be used to optimize energy consumption, monitor pollution, and improve resource management. On the other hand, the energy demands of training AI models and running large-scale AI systems could contribute to increased energy consumption if not managed properly.

Fears about AI

Expressed fears of what AI might lead to range from the relatively mundane to the apocalyptic. The more dramatic risks generally relate to AGI, not narrow AI, although the distinction is often fudged. There are serious risks, of course, even with ‘Weak’ AI.

Impact on employment and skills

Widespread use of AI could lead to mass unemployment, (although similar fears were expressed in the early years of the Industrial Revolution). This is one of the concerns of the Screenwriters Guild in their present strike. Writers wanted artificial intelligence such as ChatGPT to be used only as a tool that can help with research or facilitate script ideas, and not as a tool to replace them. The Screen Actors Guild has also expressed concern over the use of AI in their strike.

To a limited extent, they are too late. Forrest Gump, Zelig, Avatar, The Matrix, and Toy Story, all in different ways, used elements of what will need to be in any workable film AI. There has been talk for many years about using computer-generated avatars of deceased actors in live-action filming.

Military AI

Military use of AI could lead to life-and-death decisions being made by machines with no human involvement. Associated risks include the possible use of AI being used by bad actors in cyber warfare. ‘Deep Fakes’, such as those of Volodymyr Zelenskyy and Barack Obama, are only the start.

Text-based AI

I have had only limited experience with text-based AI. I have used ChatGPT to support my research into AI, but I don’t know how it would cope with longer pieces like this blog post. None of the other programs I have tried have produced anything beyond banal superficialities. Too often the output is junk. There have been reports of the AI simply making things up, which paradoxically is probably the closest it gets to originality. With the current state of development, I would not believe any publicly available AI which claimed to generate original text.

Issues for education

Even so, packages like ChatGPT seem already to be causing problems in education. Unsurprisingly, plagiarism checkers are also being deployed, although they probably use AI too! So far as I can tell, these text generators don’t provide academic references as normally used, or any link to their original sources. If developers can crack that, then there will be a real problem, not just for schools and universities.

I asked ChatGPT to list some further reading on issues around the use of AI. It gave me what looks to be a good, varied list, probably as good as anything I might find by a Google search. I asked three times, getting differently formatted results each time. Version one gave me 2 or 3 titles under a range of headings, version two gave me a list of ten titles, while version three organised the list by books, articles and online resources. There is a significant overlap, but the content varies between the three versions. None of them are properly formatted to provide references, although enough information is given to find them.

I asked again, for a list in Harvard style, and got a list of 10 books and journal articles. When asked. I got details of the requirements for Harvard Style references. A further request gave me a list of other referencing systems in use in academic circles.

Conclusions

Strong AI has been a common fear for centuries. From Pygmalion or Pinocchio to the films 2001 or AI (in itself a partial retelling of Pinocchio), similar tropes arise. I covered the same theme in my short stories, here and here.

Weak AI

The use of even weak AI raises many complex moral and philosophical questions. Some of these, such as the Trolley Problem, had been interesting ways to explore ethical issues, but now, faced with self-driving cars, they have become real. Using AI in any form of decision-making process will raise similar problems.

There is still a long way to go to get ‘intelligence.’ If it happens, I suspect it will be an accident. Eventually, though, I believe even ‘weak’ AI will be able to provide a convincing facsimile. Whether dependence on AI for critical systems is desirable is another question.

AI as a research tool

In the more limited area of text generation, ChatGPT, used as a research tool, appears to work as well as Google, without the need to trawl through pages of results. On the other hand, it is probably the later pages of a Google search which provide the equivalent experience to the chance find on the library shelves. Without careful choice of words, it seems probable that AI search tools will ignore the outliers and contrary opinions, reinforcing pre-existing biases. But then, the researcher has to actually look at that chance find on the shelf.

I asked ChatGPT to list some further reading on issues around the use of AI and to include some contrary views. This it duly did, giving 2 or 3 references under a range of sub-topics with an identified contrary view for each. As a research tool, I can see that AI will be very useful. Its use will require care, but no more care than is required in selecting and using reference material found in any other manner.

On the other hand, it seems likely that, whether we like it or not, within a few years, AI packages will be able to generate a good facsimile of original text. A good proportion of journalism is already generated in response to press releases. How long before we get AI-generated news reports created in response to AI-generated press releases about AI-developed products? All untouched by human hands…



Posted on by — Leave a comment

Thinking about size

If you go to any gallery show, the chances are you will see some huge paintings. Indeed, some are so big they might equally be described as murals. How much of that comes from the artist wanting more space in which to work, I wonder? It’s a value judgement, of course, but too often with contemporary art it seems to me to be a substitution of size for content.

Of course, simple marks can be transformed by increase in scale. Compare a single brush mark on a post card piece of paper with one drawn with full arm extension on a large canvas. In the first case, that mark is generally seen as a line. It marks off one area of the image from another without much in the way of detail. At the larger scale though, it becomes an element in its own right, it has internal structure and texture. That doesn’t mean that the huge sweeping mark is better than the small line. It’s just different. Increasing scale doesn’t, of itself, add artistic value, but it does open up the possibilities.

Contemplation

I read a quote from Mark Rothko about this. He wanted people viewing his work to enter into them.

The reason for my painting large canvases is that I want to be intimate and human. To paint a small picture is to place yourself outside your experience, to look upon an experience as a stereopticon view or with a reducing glass. However you paint the larger picture, you are in it. It isn’t something you command.

Mark Rothko https://www.markrothko.org/quotes/

He saw them as objects of contemplation, as an experience in itself rather than a record of an experience. Rothko’s paintings are indeed very large. Sitting close to them, as he wanted, it is easy to lose the world beyond the canvas. It fills your peripheral vision. After a while, the world beyond the picture fades away.

That experience is indeed a function of the scale. I tried making some versions of these works on 10” x 10” canvases. I don’t know how Rothko constructed his pictures. My approach was to build up layer after layer of colour in varying shades, wiping back, adding more colour and so on. What I got seemed to be a reasonable approximation of a Rothko surface. My relationship with it was, however, very different.

Analysis

This seems to be distinctive to Rothko, for me anyway. I had nothing similar, for example sitting in front of one of Gerhard Richter’s Cage series. To be fair, that was in a crowded show, not in a normal quiet gallery. I think there is just so much going on in the ‘Cage’ paintings that it’s easy to find oneself deconstructing them, working out how the layers lie over one another. This seems to be an analytical rather than a contemplative relationship. One of these days I intend to sit again in front of the ‘Cage’ series and see what happens.

I missed the exhibitions at the RA in London of work by Richard Diebenkorn and of the Abstract Expressionists. It is unlikely I’ll get another opportunity to see their work and make the same comparison. Watching film of Pollock at work, it is however possible to imagine achieving a similar contemplative state. His movements as he paints are like a ritual dance of some sort, with the marks and drips are forming a record of it, like some form of dance notation.

Deference

Modern art is not the only place to find the monster canvases, of course. David’s painting of “Napoleon Crossing the Alps” is about 2.6 m x 2.2 m. (There are five versions, all slightly different sizes.) Napoleon refused to sit for it, saying that the character that comes through was more important than the likeness. The resultant painting is meant to impress, to dominate the viewer. It is about Bonaparte the Emperor than Bonaparte the man.

So-called ‘History Paintings’ are all large scale. The term typically refers to any picture with a high-minded or heroic narrative, as illustrated by the exemplary deeds of its figures. The message must however be edifying and worthy of depiction.

Although perhaps not strictly speaking ‘history paintings’, the apocalyptic works of the Victorian painter John Martin also come to mind. Similarly, the revelatory, epic landscapes painted in nineteenth-century America. In all these works, the role of the viewer is almost to pay homage.

I’ve identified three distinct modes of relationship between the viewer and the very large painting; contemplative, analytical and deferential. I’m sure there are other ways of thinking about this relationship. In particular, I’m not sure if my ‘analytic’ response is driven by me being an artist myself.

I wonder though. Would a non artist react in the same way as I did to the works by the likes of Richter or would they just lose themselves in the colours and shapes?

What do you think?

Posted on by — 1 Comment

The chain of creative inspiration

This post looks at how long buried memories can resurface and trigger creative inspiration.

Albert Irvin

Back in December 2018, I went to an outstanding exhibition at the Royal West of England Academy in Bristol, Albert Irvin and Abstract Impressionism.

In 1959, Irvin visited an exhibition called The New American Painting at Tate, curated by MOMA New York and toured to eight European cities. It brought the boldest and best new artistic talent from across the Atlantic to London. The exhibition redefined what was possible for a generation of British artists.

For Irvin, it was an epiphany.

Albert Irvin and Abstract Expressionism will bring together works by Irvin and the major abstract expressionist artists that inspired him, including Jackson Pollock, Willem de Kooning, Robert Motherwell, Barnett Newman, Sam Francis and Adolph Gottlieb from UK collections and works by Grace Hartigan and Jack Tworkov on loan from the USA, giving a unique chance to see so many of these important artists’ works in this country.

RWA website

Gateshead and public art

I had come across Irvin before, sometime in the late 1980s. I was still working as a planner then, in Gateshead, with responsibility for the Town Centre. One area I was working on was improvement of the physical environment of the centre. One of the projects I identified was a dismal and uninviting roadway under a railway bridge. This underpass led from a housing estate on the opposite side of the railway line to the town centre.

My idea was to line the underside of the bridge with vitreous enamel panels. Many stations on the recently opened Tyne and Wear Metro system used this system. The photo, of Jesmond Station in Newcastle, shows one example, with artwork by Simon Butler from 1983. Making these involves screen printing the image onto the surface before firing. The panels were robust and resistant to damage. In an unsupervised location like this one, they would have been ideal.

Jesmond Metro Station, Newcastle

This is where Albert Irvin comes in. I had already had contact with the Regional Arts Association, (Northern Arts), through a minor involvement in Gateshead’s Public Art Programme. I approached them again for advice on a suitable artist. The artist they suggested was Irvin. Sadly, I never saw his work in the flesh, only some murky colour transparencies in a hand held viewer. Even more sadly for me, the project never secured funding and soon afterwards I moved to Wiltshire. Looking at Google Maps, it seems that the bridge has now been rebuilt with a rather pedestrian tiled wall. The link shows the view in 2021.

https://goo.gl/maps/kPfKeS8dQYo6Wj5MA

Since then, I’ve only been back to the Town Centre once, when I photographed the soon-to-be demolished multi-storey, ‘Get Carter’ car park. The housing estate had gone, only partly redeveloped, largely with the usual mixture of sheds found anywhere.

Photographs of the iconic 'Get Carter' car park in Gateshead, before demolition
Photographs of the iconic 'Get Carter' car park in Gateshead, before demolition
Photographs of the iconic 'Get Carter' car park in Gateshead, before demolition
Photographs of the iconic 'Get Carter' car park in Gateshead, before demolition
Photographs of the iconic 'Get Carter' car park in Gateshead, before demolition
Photographs of the iconic 'Get Carter' car park in Gateshead, before demolition
Photographs of the iconic 'Get Carter' car park in Gateshead, before demolition

Memories

I had forgotten much of this until I went to the show. Nothing I saw reminded me of those murky slides, but something triggered the memories, and they came flooding back. The comparison between the superb paintings and the tiny 35 mm slides, though, was dramatic. I was, I must admit, unimpressed by the slides, so the paintings in the show had a huge impact on me. I must have sat in front of Almada, for instance, for a good half an hour, mentally disentangling the layers, trying to work out the way they had been built up.

Another painting, Kestrel, looks much simpler, but that simplicity is deceptive. In this case how it was made is irrelevant, all that really matters are the colours, which simply sing out.

It was via this exhibition that I became aware of Irvin’s screen prints. I was lucky to see them, since they were only displayed up to Christmas. Printmaking seems always to have been an important part of Irvin’s creative practice. For me, they captured just how well he handled colour. It is to them, too, that I return for inspiration, in the form of the catalogue of his prints published in 2010 by Lund Humphries. The way he worked is well described in the book, and also seen in this YouTube video.

Irvin and screen printing

The two prints below are striking examples of his exuberant approach to colour and also capture some of the recurring motifs he uses – starbursts, quatrefoils and mock Chinese characters abound.

I’ll come back to all this in another post, looking at the way I think these and other influences have worked their way into my own work.

Posted on by — Leave a comment

Getting through a creative block (updated)

Abstract collage with block of red and orange overlain with green strips asymmetrically arranged

As the absence of posts here might indicate, I’ve been going through a period where I can’t seem to make progress, whether writing or printmaking. I’ve been casting around for ways of getting through this creative block. This post is about one of the approaches I’ve tried, inspired by artists on YouTube, Instagram and elsewhere including Jane Davies, Karen Stamper, Albert van der Zwart, and Sally Hirst whose Collage Creations course I’m currently working through. (I wasn’t when I first wrote this.)

From these, and others, I realised that a common factor in being blocked is fear. That may be fear of not getting it ‘right’, fear about wasting materials because of a lack of confidence. That’s nonsense, of course, everything doesn’t have to be brilliant every time, and nothing used with intent is wasted. Still, it’s an easy mind set to fall into, especially when you’ve been working hard. With that in mind, I set out to create quantity, not quality, setting myself some arbitrary rules to stop me overthinking.

Normally, if I get stuck I make some small gel prints, as that is a quick and effective way of getting something to look at. I couldn’t do that this time, because I felt I was in a rut. So, I created an extra disruption by choosing collage, a medium I hadn’t done a lot of work in.

I started with some very small (A6) pieces of mixed media paper and a pile of random shapes cut from prepared papers, old maps etc. I gave no thought to how they might fit together. To make the collage, the rules I set myself were:

  • work fast – no more than a couple of minutes on each. Don’t go on endless searches for the perfect shade or piece.
  • no more than 2 or 3 pieces in any one composition.
  • free form – i.e. not working to a rectangular shape. Let the paper decide

I don’t think the results are in any way finished pieces, although a few are quite pleasing. I don’t even think of them as ‘studies’. A study, to me, implies a degree of planning, of working toward something. These are ideas, no more, and like all ideas some are better than others.

I’ve put the full set so far in the slide show below.

So, what’s next? Again, I’m not entirely sure. I’ll probably make another batch in similar fashion, perhaps a little larger. I don’t want this way of working to be the new normal, so I will need to make sure I keep experimenting. I’ve started making similar pieces in a concertina sketchbook I made by folding and cutting a large A2 sheet. Working that way stops me seeing them as ‘art’, but as trials.

In a variation on this, I dripped and spattered acrylic ink across a similar A2 sheet, then folded and cut it to make another sketchbook. You can see the effect below.

EDIT: Since I first wrote this, I have enrolled on Sally Hirst’s Collage Creations course. Other factors have intervened, so my reading has got far ahead of the making. So far, though, I’m finding it worthwhile. Obviously, some bits I knew already. Sally’s clarity of exposition has, though, has enabled me to use what I knew with deliberate intent, and to build on that. I’m looking forward to working through the rest of it.

Updated and extended 28/08/2023

Posted on by — Leave a comment

What does it mean – part 3

Fourteen Discs - painting by Patrick Heron

This is the paradox that lies at the heart of the mystery of artistic creation. The meaning which a work of art has for society is not the same meaning that the artist was conscious of putting into it. This is because a work of art is not just a telephone exchange which facilitates straightforward communication. The work of art is in some profound sense an independent, live entity. It has its own life. It draws nourishment from its creator that he was totally unaware of having put into it: and it redistributes nourishment to the spectator (including the artist himself, for he is merely a spectator once the work is completed). What the work itself communicates is a transformation of all that the artist was conscious of investing in it. Its success or failure, as a transmitter of thought or emotion, simply cannot be planned and guaranteed beforehand. This is why I say that to demand a certain result from art in advance is utterly to misconceive the central creative process itself. It is to suppress spontaneity: to batten down on the subconscious.

Patrick Heron, Art is Autonomous
First published in The Twentieth Century, September 1955.
Reprinted in Painter as Critic, Patrick Heron: Selected Writings
Editor: Mel Gooding, Tate Gallery Publishing, 1998
ISBN 1-85437-258-0

More on Patrick Heron: https://www.tate.org.uk/search?q=Patrick+Heron

Posted on by — Leave a comment

Another admin update

Perelander - digital print

I’ve made a minor change on the Shop page. You can now browse everything in the shop by clicking on either ‘Medium’ or Collection. If you do, you will find several sub-categories. ‘Art under £50’, is self-explanatory. The new link, Current Work, only shows my most recent work, representative of my current practice.

Everything seems to be working correctly, but if you find any problems, please let me know, either in a comment below or by email.

Stay tuned for an announcement soon for special offers, only available to subscribers to my Mailing List.

Posted on by — Leave a comment

Changes behind the scenes

Electric avenue digital print in simulated frame

I’ve been making some changes to the shop behind the scenes, tidying up the shipping side of things. I’ve extended the range of places I will ship to, and I hope this time set the shipping costs up correctly. With luck, you won’t notice, but if you do find any errors or inconsistencies, please let me know.

Off-line, I’m also building a proper database of completed images and where they are placed. This should help me keep a better eye on things and make sure my prices are consistent. I’ve started by exporting the contents of the shop to LibreOffice. In the process, I have learned a few new tips and tricks with LibreOffice Calc (the Excel equivalent) so that’s an extra…

Posted on by — Leave a comment

Working with stripes

Abstract collage blue block with red strips

I’ve just made a set of digital images built around stripes. Working digitally is my fallback position when I’m prevented from doing anything else, whether by time, health or anything else. I had no great expectations for these. I picked on stripes for the same reason I first picked on the cross, and recently the fuji-like peak silhouette. They were a recognisable starting point.

When I posted them to my Instagram account, I made reference to Sean Scully and Johnnie Cooper, in particular the latter’s ‘Fractured Light: Johnnie Cooper, Collages 1992–1997‘. There are of course many other artists who use the stripe in their work. See here, from the Tate, for example, or here, from a US site.

I’ve been adding some of my recent digital images to the shop. I’m not sure if these will make it though. The outcome was interesting, although not quite what I’d been expecting. That’s not a problem, of course. I like these on screen, but I think they might need some physical texture to really come alive.

They started life as gel prints, which I cut up to make collage. You can see those here. To make the digital images, I used the scans of the collage made for the shop listings. I brought these back into Paint Shop Pro and then edited and recombined them in various ways.

I’m not sure of the next steps if I don’t offer them digitally. One option is going back to collage. The tissue I use to remove excess paint from the gel plate before printing would work well over solid blocks of colour, whether painted or collaged. It’s certainly a path worth exploring.

It also occurs to me that scans of the tissue could also be used in digital prints, taking the cycle round again.

Posted on by — Leave a comment

Site changes – update

Collage with simple blocks of coloured paper in green red and yellow-green

The improvements to the shop structure have now been completed. You now have several alternative options for browsing. You can choose between either the art medium used, or the subject. In addition, you can also browse through images priced below £50.

There is still some tidying up to do, but the shop seems to be functioning as it should. In particular, I am still editing, or in some cases writing, the descriptive text for each of the categories, This should also improve my presence in searches. There is some editing of product descriptions also needed, partly for style but also to clean up typos. The price structure for digital products needs reviewing, if only for consistency. Some of the prices may need adjusting to allow for the significant increases in materials costs. Finally, I need to rationalise the options available to get consistency across the board.

In a minor change, featured images in posts will from now on be clickable. Clicking will take you to the relevant product if it is in the shop or to another relevant URL.

I’m also going to be making some changes on my personal blog site, to improve links to the shop.

The next step, here, will be adding more products. I have lots of new work still not even scanned, so keep checking back. You can always sign up to my mailing list to be kept updated.

Posted on by — Leave a comment

Site changes coming up

Briseis - War Music Series - Digital Print

In January, I hope to begin restructuring the shop. It’s frankly a bit of a mess at the moment and not very logically organised, which I don’t like. I’m physically messy, I know, but mentally I tend the other way, obsessing about clarity of expression and language. That’s what makes me enjoy writing and editing, I suppose.

I have tried to avoid major restructuring up to now, but the site is now configured to automatically set up redirects, where a URL changes. This should prevent any bookmarked links from breaking. I’ll give further warnings before I make the changes.

One big change will be to do away entirely with the printmaking category. I am primarily a printmaker, after all, so finding prints shouldn’t come as a surprise. That means that the categories like Collagraph, Drypoint etc will become the first level when you open the shop.

I would also like to offer a search by image subject, landscape, townscape etc. I’m still thinking about the subjects people might find useful, and also about the changes to structures behind the scenes, that I’ll need to make. More on this later.

Comments on this aspect, or on site navigation in general, would be very welcome.

In order to help spread the word on the site in general and the work n the shop in particular, I’m going to start offering newsletter subscribers a discount voucher each year. I’m still setting that up, but don’t let that stop you signing up below or go here.

MailChimp Subscription Form

Watch this space for more news soon.