“A poet of fantastic inversions.” Poetry London

“Multifaceted, mega-fabricated, louche architecture.” Magma

“Voraciously experimental, precociously accomplished.” Poetry International

Frayed Ends: On the Longevity of Digital Poetry

Change, change — that’s what the terns scream
                                        down at their seaward rocks;
fleet clouds and salt kiss — 
everything else is provisional,
                                        us and all our works.

 —  ‘Fianuis’ by Kathleen Jamie

This paper was originally delivered as a talk at Poetry, Representation and the Archive, a symposium hosted by the University of East Anglia on 25th May 2023. My practice and research mostly concerns the amalgamation of poetry and poetry books with other forms and genres of text, including digital and interactive works, and here I want to discuss digital poetry in particular. I want to talk about the problems inherent in preserving it, and some ideas I have about how its life can be extended.

First of all, I’m going to roughly define ‘digital poetry’ as poetry which is “digital-born” (Ensslin, 2014, p.19); that is, written and designed for publication on or within a digital platform. This does not mean it is totally incapable of being transferred to or translated into a non-digital medium (I will come to talk about this in more detail) but rather that some element of the additional functionality of a digital platform is anticipated in the design. Digital poems are, therefore, poems which physically move and change, often responding to user input. They are poems which are brought into being and animated, at least in part, by computer code which an app or program (such as an internet browser) is used to interpret. There are other, overlapping terms in use – ‘e-poetry’, ‘hypertext poetry’, ‘interactive poetry’, ‘code poetry’ and so on – but ‘digital poetry’ is, at present, the most useful umbrella term for this kind of poetry.

Digital poetry is an important part of the contemporary poetry ecosystem because it is accessible in a different way to poetry which is published in print. We often talk about the concept of ‘accessibility’ as if one can simply increase or decrease the number of people to whom a particular artefact or medium is comprehensible, appealing and available. “Make poetry simpler,” we say, “in order that more people can understand it.” I would argue that in fact, the kind of alterations that would make a poem more accessible to one group of people are liable to make it less accessible to another. People become literate in using and understanding different types of media at different rates, or have access to different sorts of intellectual and technological resources. The accessibility of an artform has more to do with the plurality of forms and styles in which it is delivered than it does a particular quality of any one of those forms.

Digital poetry can be created with tools that are freely available on the internet, or which come pre-installed on most personal computers. By writing HTML, CSS and javascript code into Windows Notepad, one can quickly put together a webpage with moving, interactive text. Online resources such as Twine allow users to circumvent coding, and are accompanied by detailed manuals. Community forums and tutorial videos are easy to find with a search engine. Creators of digital poems are able to upload their creations to websites such as, through which they can either make their work available for free, viewable through an internet browser, or sell it as a downloadable file.  

Clip from ‘Wolfgirls in Love’ by Kitty Horrorshow. Created using Twine and playable through a browser window. 

Because they usually include animation and interactivity and may also make use of images, sound and ludic mechanics (for example: puzzles), digital poems appeal to different kinds of intelligence and literacy than do poems written for the page or the stage. That is not to say that they are more accessible; they are differently accessible.

I would also argue that digital poetry is an area of innovation. ‘Innovation’ is a much-abused word, and one we usually associate with progress, as if the purpose of innovation in an artform is to improve upon what already exists. My view is that innovation is best viewed as a struggle against the conservative impulse to formularise, ritualise and codify art such that success is only achievable through modest variation on (and reinforcement of) that which already exists. Sites of innovation must inevitably be sites of fluctuation, where the tools, methods and mediums of publication are constantly subject to change, resulting in a broader variety of work and a difficulty in maintaining settled criteria for its success. Inevitably, this is the kind of context that attracts the least institutional support, since such volatility increases the risk of any investment being misjudged. It should be no surprise, then, that digital poetry is ineligible for nearly all poetry prizes in this country, nor that despite the growing number of online-only poetry journals, there are few opportunities for writers of digital poetry to submit work for editorial review or curated publication.

Mary Hedger and ‘Layer Love’

Because it is thus neglected, digital poetry is at acute risk of disappearing, and I’d like to now turn to an example of such a disappearance. As part of my PhD research I coined the term ‘ludokinetic poetry’ to describe a particular kind of poem I’m interested in making and investigating. Ludokinetic poetry is not a subgenre of digital poetry, but there is a big overlap; ludokinetic poetry is defined by the use of interactive mechanisms to involve and implicate the reader in one or more outcomes of the poem. The poem here is essentially a vehicle that is operated by the reader. While this can be achieved using printed materials (in the form of a card game poem, for example), such artefacts are obviously well-suited to the digital medium.

Because ‘ludokinetic poetry’ is my own term, finding examples of it has proven difficult. Its authors might refer to their creations as ‘game poems’, ‘poetry games’, ‘interactive poems’, ‘hyper-poems’, or use any number of other terms. Casting the net as widely as I could in search of examples, I was pleased to stumble on a website belonging to a poet called Mary Hedger.

Screenshot taken from Mary Hedger’s now-defunct website using the Wayback Machine.

Outside of what I’ve learned from this website, I don’t know who Mary Hedger is. Even when I first encountered it, the site seemed largely abandoned, with no indication of anything new being added since 2014, and it seemed to have been made deliberately tricky to navigate, as if it were itself a game or a digital poem. It was mainly composed of a small number of rather cryptic images which could be clicked on to take the user to the different pages. One of these pages contained information about Hedger itself, which started out like this: 


The mauve post-its read:

“I am so, so passionate that sometimes I seem almost mad.” / “I can see flux so clearly in the world around me that my poetry can be as elusive as quicksilver.” / “I love 2 feel the bottom of every mood which can be a downer.” / “When my spirit soars, there is nothing to hold on to anywhere near me.”

No email address or social media links were supplied. At the time, what I was most interested in was the hypertext section of the website, which is subtitled “ideal medium connecting thought with poetic outcome”. In the text which accompanies each piece in this section, Hedger alternates between calling her works “hypertexts” and “poems”. Talking about ‘Layer Love’, one of the two pieces I was most interested in, she wrote:

Layer Love is a simple hypertext over three pages. It explores the ability of single words within the context of the multi-dimensional nature of hypertext, to become alive and adapt to meanings imagined by the reader when linked together under the dynamic principles of hypertext.

What then does this poem mean?

The idea is that its meanings are entirely personal, the actual words and numbers used are suggestive rather than determinative. This hints at a wider attitude of openness, lack of prejudice and freedom under the auspices of which the Internet was originally conceived.

Not long after I discovered Hedger’s website and made notes on two of the poems, it went offline. Nothing can be found now at the original address. The screenshots I have of it were taken using the Wayback Machine, part of the Internet Archive’s free digital library. The Wayback Machine takes automatic snapshots of web pages from all across the internet every few months, enabling users to go ‘back in time’ to see what a website looked like at an earlier date. Its owners say they have archived more than 800 billion web pages this way.

Unfortunately, these snapshots are not perfect. Often, as in the screenshot above, images are missing. Complex code may not work, especially if it draws on content or code libraries outside of the webpage itself. If a website was made using Flash, a now-discontinued, once-popular animation and web-building tool, it will no longer work at all on a normal internet browser, even if archived by the Wayback Machine.

‘Layer Love’ can, with some difficulty, be played through. Clicking on parts of the poem now results in a page redirection, interrupting the flow of the text. Very possibly, the whole work can be salvaged and uploaded to another website. But certainly from the point of view of the average web user it has gone from existing as an accessible ludokinetic poem, discoverable via Google, to existing in a state of disrepair, and available only to those who know the website address.

The Wayback Machine itself is under threat. On the 26th March this year a federal judge in the US found against it and in favour of Hachette Book Group, HarperCollins, John Wiley & Sons, and Penguin Random House, who are suing the Internet Archive on the basis that by scanning physical copies of books and distributing them as borrowable e-books it is engaged in making illegal derivative works.

Will ‘Layer Love’ be accessible to anyone in a year’s time? I don’t know. If it were published in a paper book or journal that had sold a hundred copies, then it has a hundred opportunities to survive in the wild. But if the Wayback Machine ceases to operate, to my knowledge there will be no copies of this poem anywhere in the world. 

There are other ways digital poems can fade into the ether. As well as the data itself becoming lost or deleted, the software or hardware that is able to parse the code may be changed and updated, and previous versions phased out, with the result that the poem can no longer be seen or experienced in the form originally intended by the author. With each new version of HTML, for example, certain tags are ‘deprecated’, which means they won’t work in modern browsers.

Additionally, the specific literacies required to operate the software or hardware, or to understand how to manipulate the poem (what rules to play by) may be forgotten, as new generations of users are brought up using different interfacing systems. The poem may also become buried; that is, it continues to technically exist somewhere on the internet, but nobody knows how to find it anymore. It’s somewhere in those several hundred billion pages that have been archived.

So what do we do?

Is the best hope for digital poetry some kind of dedicated, protected archive, run by a charitable or socially responsible institution? In fact, we all have some degree of power to aid in its preservation. We can keep private archives by saving and organising the data on our own hard drives, and we can also translate digital poems (albeit partially) into more flexible mediums. We can take screenshots, which can then be printed. On their own these will not replicate the full functionality of a digital poem, but by combining the reproduction of individual segments with a scheme, map or set of instructions that describes how these segments interact, we can increase the likelihood that a future reader is able to recreate most of the poem, even if that recreation takes place mostly in the imagination. J. R. Carpenter’s ‘This is a Picture of the Wind’, meanwhile, is an example of a digital-born text that has been successfully translated into the form of a book. Both the animation and the aleatory dimension of the original is lost, of course, but due to the fragmentary nature of the text, as well as its being framed as a kind of almanac – a work related to the changing seasons – much of the sense of volatility remains.

It may seem a simplistic suggestion, but we can also intervene by recording our own experiences of interacting with digital poems. The example I have to hand is not perfect by any means – it is an extract from my monograph, written as part of a general account of the forms taken by digital ludokinetic poems – but in the event of Mary Hedger’s site disappearing altogether and my being unable to restore the poem elsewhere, it is a useful account of the poem’s existence:

‘Layer Love’, written in 2004, begins with jumbled fragments of text, in a variety of colours and font sizes, arranged against block colour backgrounds. In large, red writing it says “over here”. In smaller, white writing it says “now here” and “where are you? come here!” Before the reader has even explored the interactive possibilities of the poem, they are implicated in it through the mode of address, made to feel as if something is expected of them. The words and phrases can be clicked and dragged, resulting in various responses. “here is no where,” the browser responds when the reader (now recruited as player) drags “now here” out of its box, and “you did that well” when “mint” is moved. Failure and success are alternately implied without the task itself, or its meaning, being specified (…) As a poem, it is deliberately open, forcing the reader to draw personalised implications from the combinations of words and the responses they prompt.”

From the same chapter, I have a longer description of another Hedger poem:

(…) “excuse” [leads] the player into a maze of textual information in which they become disorientated, fitting firmly into Caillois’ play category of ilinx by inflicting “voluptuous panic” (Caillois 1958, p.23). Once again, the poem leads with a direct address, establishing a dynamic intimacy between the author’s persona and the player:

excuseme ex c use me

ex c u se me unfinished come back please, I can never feel finished with you

The player perceives hyperlinks in the first page, embedded within the stuttering text: “no?”, “excuse” and “e-sEx”, the latter of which acts like the discovery of a hidden message (“e-sex”) within the litany of repeated variations of “excuse me”. On a thematic level, the piece invokes romantic obsession (“i want – i want – i want – i want”) and agitation over the appropriateness of a relationship; if the player opts for “no?” then a different, more authoritative voice enters from the right-hand side of the screen, saying “it’s right to think about consequences … she’s married”. But whichever route the player chooses to take, they will end up on a page where segments of text overlap, literally talking over one another (the word “hello” is repeated, its font size gradually increasing, over the top of “for a while must not fantasise about o at night before i go to sleep”) and where clicking or moving the cursor may result in interruptions from further text boxes, further information or enquiry. “is this what you said yes to?” demands one, if the player hovers over “you know how i feel you know how i think”. Implicated in the drama of the poem by way of the second person and the conspiratorial immediacy of its voice, the player must now try to find their way to the end.

In so doing, they enact the process of resolving (or otherwise quieting) the contradictory impulses represented by the individual segments of text talking over one another. The nature of the resolution is left open; the final link is a large letter Y following a passage that reads:

yes, it’s true, i am unable to
yes it’s impossible
yes, i can’t

The persona is cut off mid-sentence. On the other hand, the word “yes” has taken over from “excuse” as the word that is repeated most fitfully, with lines on the final page that hint at a coming to terms with desire: “yes, i am out of control … yes i want what i want”.

I’d like to end with a general, wide-ranging recommendation that we look on all our written reports and responses to poetry as a form of preservation – or rather, conservation. One of the biggest threats to our ability to make present-day poetry accessible to future generations is the pervasive myth that a poem is an object to be admired, rather than a component in a living circuit that is completed by the engaged reader. If we withdraw entirely into habits of cursory evaluation – applause, genuflection, comparison, dismissal – then we rob those future readers of valuable context. The work that is most likely to survive intact is the work that supplies, or becomes part of, an ongoing conversation, so that someone encountering the frayed ends of that conversation many years from now can follow those threads back to the artefact that gave rise to them, and find it still humming away.


Caillois, R. Man, Play and Games. Chicago: University of Illinois Press, 1958.
Ensslin, A. Literary Gaming. Cambridge, Massachusetts; The MIT Press, 2014.
Hedger, M. excuse. 2004. Accessed July 10, 2018. pages/hypertext%20poetry.htm.
Hedger, M. Layer Love. 2004. Accessed July 10, 2018. hypertext%20poetry.htm.
Stone, J. 2022. Dual Wield: The Interplay of Poetry and Videogames. DeGruyter Oldenbourg.