• 38 Posts
  • 148 Comments
Joined 7 years ago
Cake day: January 21st, 2020
  • leanleft@lemmy.mltoGlobal News@lemmy.zip•'Nobody seems to care': Russian city shrugs off plague reportsEnglish
    1·
    9 days ago

    summary from date-10/7

    ## Irkutsk laboratory death and public-health response

A young female employee of the Irkutsk Anti-Plague Research Institute in Siberia died after being hospitalized with severe pneumonia, prompting concerns that she may have been exposed to a dangerous pathogen during laboratory work. Early reports said she had told colleagues that she accidentally broke a test tube containing live bacteria while collecting samples, although Russian authorities later said their inspections had found no confirmed laboratory accident involving pathogenic microorganisms.

The death triggered an extensive precautionary response. Regional health authorities identified and medically monitored roughly 200 contacts, while some hospital units reportedly limited admissions. A nearby aluminum plant also instructed employees to wear masks amid uncertainty and conflicting reports. Artem Fominikh, director of the Irkutsk Aluminum Plant, urged calm and said that unverified information was circulating.

The Irkutsk Anti-Plague Research Institute—formally the Irkutsk Antiplague Research Institute of Siberia and the Far East—was established in the 1930s to combat plague and later broadened its work to other high-consequence infectious diseases. It plays a public-health role in disease surveillance, laboratory diagnosis, prevention, and research across Siberia and Russia’s Far East, including regions where plague can occur naturally among wild animals and their fleas.

At the center of the concern was the possibility of **pneumonic plague**, a severe lung infection caused by the bacterium *Yersinia pestis*. Russian officials did not initially confirm the cause of death, saying laboratory and tissue testing was still underway. The head of neighboring Buryatia, Alexei Tsydenov, said the worker may have died from plague, while also stressing that there was no recognized plague outbreak in his region.

## Official assessment and containment measures

Russian health authorities, including Rospotrebnadzor, said they implemented a full set of preventive measures: identifying contacts, conducting medical observation, testing people who had been exposed, and monitoring the local epidemiological situation. Officials reported that no dangerous infectious pathogens had been detected among the contacts. Later updates indicated that monitoring had been completed for all identified contacts, with no confirmed secondary cases.

According to the reported Russian findings:

- Some contacts tested positive for common respiratory infections, including COVID-19 and rhinovirus
- No evidence emerged that a dangerous pathogen had spread from the deceased worker to close contacts
- Inspections reportedly found no confirmed incident involving pathogenic microorganisms at the institute
- Testing of the deceased reportedly did not identify microorganisms associated with her professional work, though the full diagnostic picture was initially still being clarified
- Authorities said the regional sanitary and epidemiological situation was stable and under control

The World Health Organization said it was aware of reports of a laboratory worker in Irkutsk Oblast dying from severe pneumonia on October 2, 2026. WHO stated that the cause of death had not yet been officially confirmed and that testing was underway. It noted that all known contacts had been identified and monitored, with no symptoms reported among them at that stage. WHO’s initial risk assessment characterized the threat as moderate to low for the immediate Irkutsk area, low for Russia overall, and very low for the broader WHO European Region. 

WHO also said it was communicating with Russian authorities under Articles 9 and 10 of the International Health Regulations (2005). That process involves verifying the event, reviewing laboratory findings, understanding why contact tracing was undertaken, and assessing whether the incident could have international public-health implications. 

The European Centre for Disease Prevention and Control likewise reported no known secondary cases and no evidence of sustained person-to-person transmission, while emphasizing that its assessment rested on limited available information. 

## Why pneumonic plague caused concern

Plague remains rare, but it is taken seriously because of its severity and because its pneumonic form can be transmitted through infectious respiratory droplets during close contact. It is also one of the pathogens long discussed in biodefense planning because it could, in principle, be deliberately aerosolized.

There are three principal clinical forms of plague:

| Form | Main route or feature | Key concern |
|---|---|---|
| Bubonic plague | Usually follows the bite of an infected flea | Fever and painful swollen lymph nodes, known as buboes |
| Septicemic plague | Bloodstream infection | Can progress rapidly to shock and organ failure |
| Pneumonic plague | Infection of the lungs | Can spread between people through close-range respiratory droplets |

Pneumonic plague is the least common but most dangerous form. It can arise when someone inhales infectious *Y. pestis* particles, or when untreated bubonic or septicemic plague spreads to the lungs. A person with pneumonic plague may infect close contacts through droplets expelled while coughing. However, ordinary spread requires close and usually sustained exposure; the bacterium is not generally thought to behave like a highly airborne virus that remains broadly suspended throughout indoor spaces for long periods.

That distinction matters. A single suspected case—especially one isolated in a medical setting—does not automatically signal a wider outbreak. Rapid identification, isolation, contact tracing, monitoring, and preventive antibiotics for eligible close contacts can sharply reduce the risk of onward transmission.

## Laboratory safety and transparency

The case drew particular attention because it involved a worker at a laboratory studying high-risk pathogens. Laboratory-acquired infections are uncommon but possible wherever dangerous microbes are handled. They can result from accidents such as sharps injuries, broken containers, aerosol-generating procedures, failures in protective equipment, or breakdowns in routine biosafety practices.

Experts quoted in reporting emphasized that a suspected occupational exposure to plague would normally lead to an immediate risk assessment and, where appropriate, prompt post-exposure antibiotic prophylaxis. The public does not yet have a complete account of the worker’s medical timeline, the material she may have handled, whether an exposure occurred, or what preventive treatment she received. Those questions are important, but they cannot be answered reliably until Russian authorities release final laboratory and investigative results.

The episode also illustrates why transparency is central to public confidence. When officials rapidly communicate what is known, what is uncertain, what testing has been done, and whether transmission has occurred, they reduce rumors and help neighboring countries calibrate precautions appropriately. U.S. officials, including President Donald Trump and Secretary of State Marco Rubio, called for greater information-sharing, while Russian officials said control measures were in place and the situation was stable.

Several neighboring countries increased preparedness measures. Uzbekistan said it had raised readiness for highly infectious diseases, and Kyrgyzstan announced enhanced quarantine controls at border crossings. Such measures reflect caution rather than proof that international spread occurred.

## Treatment and prevention

Plague is treatable with antibiotics, but timing is critical. Untreated pneumonic and septicemic plague can be rapidly fatal, whereas early treatment can substantially improve outcomes. Management may include antibiotics, respiratory and circulatory support for severe illness, isolation of pneumonic cases, and preventive antibiotics for people with significant close exposure.

Key public-health measures include:

- Immediate clinical assessment of anyone with compatible symptoms and relevant exposure
- Isolation and appropriate infection-control precautions for suspected pneumonic plague
- Prompt laboratory testing through qualified public-health systems
- Identification and monitoring of close contacts
- Preventive antibiotics when public-health authorities determine exposure risk warrants them
- Clear, evidence-based communication to prevent both complacency and panic

For the public, the available information did not support a conclusion that a plague outbreak was occurring in Irkutsk or that there was sustained person-to-person spread. The reported absence of secondary cases, completion of contact monitoring, and lack of dangerous-pathogen findings among contacts were reassuring signs. At the same time, the death of a laboratory worker handling—or potentially handling—dangerous pathogens underscored the continuing importance of rigorous biosafety, biosecurity, transparent investigations, and rapid international notification when a potentially serious infectious event occurs.
  • leanleft@lemmy.mltoNews@lemmy.world•Kyrsten Sinema’s former bodyguard and ex charged after rampage through her homeEnglish
    1·
    15 days ago

    title: “Sinema’s ex-lover had ‘large’ knife, threatened to ‘take her down’ before arrest at Cheesecake Factory: police” url: “https://www.fox10phoenix.com/news/sinemas-ex-lover-had-large-knife-threatened-take-her-down-before-arrest-cheesecake-factory-police” author: “Elias Weiss”

    Kyrsten Sinema’s ex-bodyguard, beau re-arrested

    Kyrsten Sinema’s ex-bodyguard Matthew Ammel was arrested again in Phoenix after violating pretrial release.

    The former bodyguard and ex-boyfriend of one-time Arizona Sen. Kyrsten Sinema was arrested again on Sept. 30 after authorities say he violated his pretrial release by dropping off a “large, military-style knife” at a gun store — and threatening to kill judges and police officers.

    What we know:

    Prosecutors with the Maricopa County Attorney’s Office filed an emergency motion on Wednesday requesting a judge increase the bond for 39-year-old Matthew Ammel to $100,000 cash-only. The filing comes less than two weeks after Ammel’s initial arrest for allegedly causing nearly $200,000 in damage to Sinema’s Cave Creek home, when he was given supervised release.

    Now, state attorneys are warning that the U.S. Army veteran poses an escalating danger to the public and to Sinema after threatening to “take her down.”

    Matthew Ammel appears in court, center, following his re-arrest for violating pretrial release conditions in a felony domestic violence case involving former Arizona Sen. Kyrsten Sinema, right.

    According to court records, Pretrial Services was alerted by North Carolina authorities on Tuesday that Ammel was actively trying to acquire guns and tactical gear while making threats against a judge. Later that afternoon, Ammel’s court-ordered GPS tracker placed him at a firearm retailer in north Phoenix, where a sheriff’s detective learned Ammel had left a large “hunting-style” knife to be engraved.

    State prosecutors said in court documents that police have serious safety concerns due to Ammel’s “erratic” behavior, “highly skilled” military training and efforts to have “military friends” help him assemble a rifle. Court filings say Ammel prepared for violence by discussing funeral arrangements and making threats against the lives of sitting judges, lawyers and police personnel.

    The arrest:

    Detectives with the Arizona State Gang Task Force helped lure Ammel out of his hotel room on 24th Street and Camelback Road and asked him to walk across the street to The Cheesecake Factory parking lot, where he was taken into custody on Sept. 29 at 9:45 p.m.

    Ex-bodyguard accused of trashing former Arizona Senator’s home

    The former bodyguard to ex-Arizona Senator Kyrsten Sinema is facing a felony charge after deputies say he caused over $200,000 in damage at her Cave Creek home. FOX 10’s Jacob Luthi reports.

    Dig deeper:

    Prosecutors now say Ammel also broke a court protective order shortly after his initial release by getting an acquaintance to reach out to Sinema on his behalf. On top of his Arizona charges, Ammel was already serving unsupervised probation for a prior felony case in North Carolina.

    The backstory:

    The original criminal charge stems from an Aug. 30 incident where deputies say Ammel entered Sinema’s home while she was away, demanding she post a video to X. When Sinema refused, Ammel allegedly destroyed 22 paintings valued at over $123,000 by throwing them into her backyard pool and shattered expensive alcohol bottles and mirrors throughout the house.

    Ammel was initially taken into custody on Sept. 17 after he was discharged from a mental health facility; he was charged with felony aggravated criminal damage domestic violence. The criminal cases follow Sinema’s court filing acknowledging an affair with Ammel in response to an “alienation of affection” lawsuit brought against her in North Carolina by Ammel’s wife.

    What’s next:

    A judge set a $25,000 bond following the knife incident. Ammel will be arraigned on Oct. 7, with a pretrial conference scheduled for Jan. 20, 2027.

    Map of where a state task force nabbed Sinema’s ex a second time:

    The Source: Information in this report was gathered from Maricopa County Superior Court records, the Arizona Department of Public Safety, and North Carolina civil and criminal court records.

  • leanleft@lemmy.mltoTechnology@lemmy.world•Americans Are a Lot More Worried About AI Than the Rest of the WorldEnglish
    6·
    22 days ago

    my theory: i feel like the first wave of any product is overhyped and has less accountability to deliver.

    "Hoverboards (2015) The “self-balancing scooter” became a global sensation almost overnight, with demand skyrocketing across retail and online markets. Because manufacturers rushed to meet this sudden explosion in demand, quality control was ignored. This led to a wave of defective lithium-ion batteries that frequently overheated and caught fire, resulting in massive recalls and bans in several countries.

    The Segway (2001) Before its release, the Segway was shrouded in extreme secrecy and hype, with some predicting it would “change the way cities are built.” When it finally launched, the reality didn’t match the excitement; it was too bulky for sidewalks, too slow for roads, and far more expensive than consumers were willing to pay for its actual utility. "

    straight off the top of my head… Blu-ray . everyone said it was better. it triggered the need to have “the latest greatest tech” but now… i dont even know how that would have benefited me, personally. it would have been more rational if they had a more tangible benefit… like a smaller disk.

    also gameboy. …however i cant think of any criticism aside from costing alot.

  • leanleft@lemmy.mlOPtoLocalLLaMA@sh.itjust.works•Code2LoRA: Hypernetwork-Generated Adapters for Code Language Models under Software EvolutionEnglish
    22·
    24 days ago

    Speculative Ideal Use-Cases for “Hypernetwork-over-Diffs” Adaptation

    First: the suitability checklist

    Extracting what actually made Code2LoRA-Evo work, an ideal candidate domain needs all of these properties:

    Property (from the paper) Why it matters
    1. Many distinct “instances” (repos) with shared structure Hypernetwork amortizes across breadth; gains saturated at ~200 repos — you need at least dozens–hundreds of instances
    2. Each instance has its own conventions/symbols The task must reward instance-specific knowledge, not generic skill
    3. The corpus drifts in bursts Bursty edits (Fig. 2) make snapshots structurally inadequate; a calm corpus → static adapter suffices
    4. Short, structured targets The task was 3-token assertion completion; the 2048-dim repo bottleneck supports symbol/convention recall, not long-form synthesis
    5. Query volume / latency sensitivity Zero inference tokens only pays off at high query rates or constrained contexts
    6. Updates are cheap diffs The GRU trick works because a diff ≪ re-encoding the corpus

    Tier 1 — Near-term, high-confidence (code-adjacent)

    Same data substrate, different tasks. Diffs already exist in git history, so the pipeline transfers almost intact:

    • Test generation & mutation testing — the natural mirror image of assertion completion: predict the assertion itself from implementation. Evolves with every commit; conventions (fixtures, mocking idioms) are repo-specific.
    • Code review assistants — given a diff Δt, predict review findings. The GRU-over-diffs structure is literally the input: reviewer style and repo standards drift as the codebase matures.
    • Bug triage / root-cause localization — map a failing test to the offending module. Requires deep knowledge of the repo’s architecture, which evolves.
    • Migration & upgrade assistants (framework version bumps, API deprecations) — the knowledge base is the changelog; bursty by nature.
    • IaC / config-as-code (Terraform, Kubernetes manifests) — conventions are fiercely org-specific, repos evolve constantly, targets are short and structured (a resource field, a value). Arguably more convention-bound than Python source.

    Tier 2 — Mid-term: institutional knowledge bases that drift

    Here “repository” becomes “an organization’s living corpus,” and “commit” becomes “revision.”

    • Internal engineering wikis / runbooks (SRE)

      • Instance: one company’s or one service’s runbook set. Drift driver: every postmortem patches the playbook.
      • Task: given an alert signature, complete the next remediation step — short, structured, convention-heavy (“in our stack, escalate to team X via channel Y”).
      • Fit: bursty updates after incidents; per-service identity; high query rate during incidents → token overhead genuinely hurts.

    • Contract drafting within a firm’s clause library

      • Instance: a law firm’s playbook of standard clauses. Drift driver: negotiation outcomes and regulatory changes update preferred wording.
      • Task: clause-slot completion (“governing law shall be ___”, limitation of liability caps) — short targets, firm-specific conventions.
      • ⚠️ But see the auditability caveat below — this domain may demand RAG despite its costs, precisely because you must cite the source clause.

    • Clinical pathways & hospital protocols

      • Instance: one hospital system’s formulary + treatment protocols. Drift driver: formularies, dosing guidelines, and pathways are revised continuously.
      • Task: complete order-set fields, flag protocol deviations — short structured outputs, deeply institution-specific.
      • Fit is strong mechanically; the stakes amplify both the staleness problem (a stale adapter is dangerous) and the value of cheap updates.

    • Versioned product documentation for support agents

      • Instance: one SaaS product’s docs, versioned by release. Drift driver: every release.
      • Task: answer “what does feature X return / default to in this customer’s deployed version” — the version axis is a natural diff stream, and per-deployment adapters beat one giant context.

    • Compliance & accounting rulebooks

      • Instance: one firm’s policy manual under evolving standards (IFRS updates, regional regulation).
      • Task: complete the applicable threshold/rule given a transaction profile.
      • Bursty (regulations land in waves), instance-specific, short targets.

    Tier 3 — Speculative but structurally perfect

    • Scientific lab notebooks / protocols — each lab’s evolving protocol corpus; targets like reagent concentrations or parameter values; updates burst around experiments.
    • Personal/enterprise style-guide writing assistants — the “repo” is a brand voice guide + past documents; it drifts as the brand evolves; targets are short lexical choices (tone, terminology). Beware: corpus may be too small for the 2048-dim bottleneck to be stressed — the adapter might be trivially learnable, reducing the method’s edge over simple fine-tuning.
    • Live dashboards of organizational knowledge — adapter geometry (the paper’s t-SNE / per-repo EM analysis, §F.4) could double as a drift monitor: if a repo’s adapter wanders far from its cluster or its EM drops along the commit axis (Fig. 9), that’s a signal of documentation decay or anomalous change. This generalizes to “knowledge-base health monitoring” in any Tier-2 domain.

    Where this approach would fail — anti-patterns

    1. Audit-critical tasks requiring citation. Parametric knowledge can’t be cited. Legal reasoning, medical decisions, and regulated finance often require a retrievable source span — RAG’s token cost is the price of attribution. Code2LoRA’s own risk section flags attribution amplification; in these domains that’s disqualifying unless paired with a citation layer.
    2. Long-form generation targets. The evidence base is short-value completion (median 3 tokens). Whether a 2048-dim corpus embedding can steer multi-paragraph generation is unproven and doubtful.
    3. Few-instance domains. The hypernetwork needs breadth to amortize (≥ ~dozens of instances). A single organization’s lone corpus can’t train the head — per-corpus LoRA is simply better there.
    4. Slow-drift corpora. If the knowledge base changes yearly, snapshot adapters never go stale; the GRU adds cost without benefit. The bursty-commit phenomenon is the method’s reason to exist.
    5. Verbatim-recall tasks. Compressing millions of tokens into 2048 dims destroys exact strings. Tasks needing literal quotes (statutory text, exact dosages) will be unreliable without a retrieval backstop.

    The generalization thesis, in one sentence

    Any setting where (a) many entities each maintain a large, convention-laden corpus, (b) that corpus evolves in bursts, © downstream queries are high-volume with short structured answers, and (d) citation is not mandatory — is a candidate for “X2LoRA”: freeze the backbone, generate the entity-adapter in one pass, refresh it one diff at a time.

    Code is simply the first domain where the diff stream (git) already exists, is machine-readable, and is free — which is why the paper could build it today. Docs, protocols, and rulebooks have the same structure but lack the free diff infrastructure; their adoption hinges on building the equivalent of git for their corpora first.