HEDL u data strutturata mingħajr nefħa ta' JSON
The invoice hiding in your braces
JSON won because it is boring in exactly the right way. Humans can read it. Machines can parse it. Every language has a library for it. If two systems need to exchange an object and nobody wants a standards meeting, JSON is usually where the conversation ends. Fine. There are worse compromises. Many of them have enterprise in the name.
The problem is not JSON as a web format. The problem is what happens when we push JSON into language-model workflows andpretend the cost is free. A model does not see a tidy object in the way an application parser sees one. It sees tokens. It reads the same keys again and again. It spends context on punctuation, repeated field names, wrappers, nested scaffolding, and shape reminders that were already known before the first record arrived.
That waste used to be mildly irritating. With AI systems, it becomes a product problem. Every repeated key competes with evidence, instructions, examples, citations, and actual user content. Every redundant structural token is a little tax on the useful work. The invoice does not say needless braces, because invoices lack poetry. It says tokens.
HEDL starts from a plain observation: when the schema is known, repeating the schema inside every record is silly. Declare the structure once. Encode records positionally. Keep the semantics exact. Convert back to the formats existing systems already expect. That is not anti-JSON ideology. It is a refusal to pay the model to reread the same road sign every ten metres.
This matters because structured AI work is not just chat. It is extraction, classification, tool calls, data transformation, review packets, evidence bundles, MCP calls, workflows, and agents passing objects to each other all day. The more serious the system becomes, the more structure it needs. If structure is expressed in the most verbose possible way, the system pays for its own discipline.
JSON is not the villain
It would be easy, and lazy, to write this as a JSON complaint. JSON has real strengths. It is ubiquitous, debuggable, easy to pipe through existing tools, and good enough for a huge amount of application work. The point is not that JSON is bad. The point is that JSON is often used in places where the receiving side already knows the shape, and there the repetition stops being clarity and starts being cargo.
Consider a structured extraction task. The schema says every answer has a name, source, value, confidence, and rationale. Now imagine sending hundreds of rows to a model or receiving hundreds of rows back from one. JSON repeats those field names for every object. The application parser does not mind. The model context does. The context window becomes a deliveryvan full of labels instead of goods.
HEDL treats the schema as a contract. It names the fields and types once. The records then carry values in order. That sounds obvious because it is. Many efficient formats have made similar tradeoffs for decades. The difference is that HEDL is aimed at LLM-facing structured workflows where human debuggability, conversion, and tool compatibility still matter. It is not a binary blob lobbed over a wall with a note saying good luck.
L-implimentazzjoni tal-HEDL tinkludi appoġġ għall-librerija Rust, użu permezz ta' CLI, uċuħ ta' MCP server u proxy, WASM, FFI u bindings, kif ukoll konverżjoni minn u għal formati komuni. Dik il-kombinazzjoni hija importanti. Format għal flussi tax-xogħol tal-AI ma jistax ikun biss kompatt. Irid jidħol u joħroġ mid-dinja mħawda mingħajr ma jsir kult privat. L-APIs eżistenti għad iridu JSON. In-nies għadhom jispezzjonaw id-dejta. L-għodod għad iridu round trips. Il-format irid ikun dens mingħajr ma jsir antisocjali.
L-istruttura bħala kuntratt
Il-biċċa l-kbira tal-fallimenti tal-AI madwar dejta strutturata mhumiex drammatiċi. Huma żgħar, u dan jagħmilhom aktar diffiċli biex jittieħdu bis-serjetà sakemm jiswew flus veri. Kamp jitbiegħed. Valur jitqiegħed taħt iċ-ċavetta ħażina. Mudell jarmi oġġett plawsibbli b'qasam fakultattiv nieqes. Parser jaċċetta forma li kellu jirrifjuta. Għodda downstream tirċievi kważi l-ħaġa t-tajba, l-aktar tip ta' ħaġa perikoluża fis-softwer.
L-approċċ tal-HEDL ibbażat fuq l-iskema l-ewwel huwa utli għax jagħmel l-istruttura espliċita qabel ma r-rekords jibdew jiċċaqilqu. L-iskema mhijiex suġġeriment laxk ta' prompt. Hija l-ħaġa li tgħid lill-qarrej kif jinterpreta l-valuri. Ir-rekord huwa kompatt għax m'għandux bżonn jirrakkonta lilu nnifsu ripetutament. L-għodod tal-madwar xorta jistgħu jivvalidaw, jikkonvertu, u jgħaddu dejta lil sistemi li jippreferu JSON, YAML, XML, CSV, jew forom konvenzjonali oħra.
Ir-round trips huma t-test ta' jekk format huwiex ta' għajnuna jew sempliċiment għaqli. Jekk JSON jidħol, HEDL jimxi permezz tal-fluss tax-xogħol, u JSON joħroġ bl-istess semantika, is-sistema tikseb densità mingħajr ma titlef il-kompatibbiltà. Jekk it-tifsira tintilef fis-skiet, il-format fela. L-imġiba t-tajba taħt pressjoni mhijiex li tgħolli l-ispallejn u tgħaddi l-oġġett downstream. Hija li timblokka, tirrapporta, u ġġiegħel l-ambigwità toħroġ fid-dieher.
Hawnhekk il-HEDL joqgħod tajjeb ma' bqija tal-istack Dweve. Ledger jieħu ħsieb li l-avvenimenti operazzjonali jibqgħu ispezzjonabbli. AION jieħu ħsieb li l-provi tad-deċiżjonijiet jistgħu jiġu ċċekkjati. Trace jieħu ħsieb li l-komputazzjoni tista' terġa' tintlagħab. HEDL jieħu ħsieb li dejta strutturata tista' tiġi rappreżentata b'mod dens u restawrata eżattament. Dawn ix-xogħlijiet imissu lil xulxin, iżda mhumiex l-istess xogħol. Għal darb'oħra: inqas kliem sħun u vagi, aktar konfini utli.
Il-benchmark mhuwiex dekorazzjoni
It-talbiet dwar il-prestazzjoni madwar l-infrastruttura tal-AI spiss jinkitbu bħal stejjer tas-sajd. In-numru jikber kull darba li jingħad mill-ġdid. HEDL għandu talba konkreta ta' benchmark: 571 kompitu ta' estrazzjoni strutturata fuq seba' datasets, 56 fil-mija inqas tokens minn JSON, u qligħ ta' 10.3 punti perċentwali fl-eżattezza fuq JSON.
Dawk in-numri għandhom jinqraw bħala talba ta' benchmark, mhux bħala liġi universali tal-fiżika. Huma jiddeskrivu setup ta' benchmark. Ma jfissirx li kull fluss tax-xogħol b'mod maġiku jikseb l-istess riżultat. Iżda jispjegaw għaliex il-format jeżisti. L-għadd tat-tokens mhuwiex nota f'qiegħ il-paġna dwar l-implimentazzjoni f'sistemi LLM. Huwa parti mill-interface. Jekk żewġ rappreżentazzjonijiet iġorru l-istess tifsira u waħda taħraq ħafna aktar kuntest, dik itqal mhijiex newtrali.
The accuracy gain is especially interesting. It suggests the benefit is not only cheaper prompts. A cleaner representation can also make the task easier for the model. That should not be surprising. If the model spends less attention on repeated syntactic clutter, it has more room for values and relations. This is the same reason good forms beat messy forms in human work. The human may be smart, but do not hand them a tax form written by a printer having a difficult childhood.
There is a broader design lesson here. AI interfaces should not be judged only by whether the model can cope. Models can cope with many bad interfaces. People can also carry furniture up stairs with poor grip and no plan. That does not make it architecture. A good AI interface reduces avoidable work, exposes structure, preserves meaning, and fails loudly when the structure is wrong.
Why proxy surfaces matter
A format rarely wins by being pure. It wins by fitting the ugly middle. HEDL's MCP and proxysurfaces matter because most organisations cannot simply announce that everything now speaks a new representation. They have existing APIs, data stores, validation rules, dashboards, notebooks, and export formats. Replacing all of that to save tokens would be aheroic way to lose friends.
The proxy pattern is more practical. Let models and tools benefit from dense structured representation where it matters. Convert at the boundary. Validate before data leaves the controlled path. Keep downstream JSON compatibility. Let systems that expect JSON receive JSON, but stop forcing the model to haul the full JSON shape through every internal step.
This is also where governance enters, quietly and usefully. If the proxy validates structure, it can reject malformed objects before they become business facts. If it preserves a lossless round trip, it can prove that conversion did not change the meaning. If it keeps compatibility with existingsystems, it can be adopted without turning every integration into a migration programme. We are European. We have enough migration programmes. Some of them still have steering committees from 2014.
For agent systems, the proxy is even more important. Agents pass structured calls and results across boundaries. They call tools, receive outputs, update memory, produce artifacts, and hand state to other agents. A dense representation with validation keeps those handoffs less wasteful and less ambiguous. It does not make the agent wise. It makes the envelope less stupid. That is a respectable engineering outcome.
Where HEDL should not be used
Every useful tool has a place where it should not be used. HEDL is not a replacement for every JSON file on earth. If asmall config file is read once by a human and edited twice a year, JSON or TOML will survive the tragedy. If a public API needs maximum familiarity and the payloads are tiny, JSON is fine. If the schema is genuinely unknown and ad hoc, schema-once encoding is not the right starting assumption.
HEDL becomes interesting when structure is repeated, volumes are meaningful, model context is expensive, round trips matter, and tools need compatibility at the edges. That is why LLM workflows are such a good fit. They sit precisely at the intersection of structured intent and token economics. They also tend to grow from prototype to production faster than anyone planned, because apparently nobody has learned this lesson despite the entire history of software looking mildly offended in the corner.
It-triq prattika għall-adozzjoni għandha għalhekk tkun dejqa. Tiktibx l-organizzazzjoni mill-ġdid. Agħżel fluss tax-xogħol ta' estrazzjoni strutturat. Agħżel mogħdija ta' sejħiet ta' għodod tal-aġent. Agħżel fruntiera ta' prokura MCP. Kejjel l-użu ta' tokens, ir-rata ta' falliment, ir-riżultati tal-validazzjoni, u l-fedeltà ta' round-trip. Jekk iċ-ċifri jżommu, espandi. Jekk ma jżommux, żomm il-ħaġa ta' dejjem. L-għan mhuwiex li tadura format. L-għan huwa li tieqaf tħallas għal struttura li tista' tiġi evitata.
Il-lezzjoni
Il-lezzjoni ta' HEDL hija li l-istruttura mhijiex b'xejn sempliċement għax hija utli. Fis-softwer ordinarju, ċwievet ripetuti huma l-aktar ta' fastidju. Fi flussi tax-xogħol tal-AI, huma kuntest, flus, attenzjoni, u wiċċ ta' falliment. Jekk l-iskema hija magħrufa, li tirrepetiha f'kull rekord spiss hija l-inqas għażla immaġinattiva disponibbli.
HEDL jagħmel skambju sempliċi: iddikjara l-istruttura darba, ikkodifika r-rekords b'mod dens, ippreserva s-semantika, ikkonverti lura meta jkun meħtieġ, u ivvalida fil-fruntiera tal-prokura. Mhuwiex sostitut għal JSON bħala l-lingwa komuni tal-web. Huwa envelop intern aħjar għal xogħol AI strutturat fejn il-mudell għandu jqatta' l-kuntest tiegħu fuq it-tifsira, mhux fuq il-qari tal-istess ismijiet ta' kampijiet sa tmiem il-baġit.
Dak huwa l-istandard utli għall-infrastruttura tal-AI. Mhux intelliġenza għall-fini tagħha stess. Mhux taxxa tan-novità. Mhux format li jeħtieġ li kulħadd ibati. Saff tajjeb ineħħi l-iskart, iżomm il-kuntratt espliċitu, u jħalli l-bqija tas-sistema tkompli taħdem. HEDL jaqla' postu meta l-oġġett isir iżgħar mingħajr ma t-tifsira ssir aktar dgħajfa.
JSON jista' jibqa'. Ma għamel xejn ħażin. Jista' saħansitra jixrob kikkra kafè. Sempliċement m'għandux bżonn joqgħod ġewwa kull sejħa tal-mudell iġorr l-istess sett ta' ċwievet bħal raġel li jiċċaqlaq id-dar mgħarfa waħda kull darba.