<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Turn &amp;mdash; alexandriadickey</title>
    <link>https://blog.0yu.de/alexandriadickey/tag:Turn</link>
    <description></description>
    <pubDate>Fri, 20 Nov 2026 16:47:51 +0100</pubDate>
    <item>
      <title>Building a Useful Delivery Risk Register: AI development services</title>
      <link>https://blog.0yu.de/alexandriadickey/building-a-useful-delivery-risk-register-ai-development-services</link>
      <description>&lt;![CDATA[pteams building customer and employee assistants often approach AI development services through questions about voice and conversational interaction design. In Building a Useful Delivery Risk Register, A conversational interface must manage recognition errors, interruptions, context, identity, tool calls, Resource: enterprise generative ai development services and user expectations in real time. ai software development Company A risk management brief must resolve which uncertainties require mitigation, acceptance, transfer or a stop decision. For an owned and testable risk register, Discover: What does ai company do? search language such as &#34;conversational ai development services&#34; supplies context for that decision, not evidence that one option is universally suitable./pp##Turn related queries into accountable questionsInterest in &#34;generative ai development services company&#34;, &#34;ai development services company voicebot development services&#34;, &#34;top ai developer companies&#34;, &#34;ai development services company voice bot development services&#34;, and &#34;generative ai app development services&#34; creates several entry points to risk management. Reviewers can connect those entry points to explicit limits, observable behavior Explore: ai development consulting and a correction path inside an owned and testable risk register. The resulting owned and Explore: Explore: adaptive ai development services ai proof of concept development services testable risk register record explains what is known, what remains uncertain and which event should reopen the decision./pp##Write risks as observable conditionsAn owned and testable risk register keeps the risk management discussion reviewable. The source topic states this practice: ai development pros and cons In Building a Useful Delivery Risk Register, Conversation design should define intents, turn handling, confirmation, AI software development company repair, escalation, privacy notices, latency, and session state. A connected practice comes from generative system design and controlled outputs: For an owned and testable risk register, Design should separate instruction, context, generation, validation, citation, Details: ai voice agent development services and user correction into observable steps. Together they define what happens before commitment in risk management and what remains in an owned and Read more: generative ai development services testable risk register after the decision./pp##Set failure boundaries for risk managementThe primary risk record says: For an owned and Discover: ai as a service companies testable risk register, A fluent response can conceal misunderstood input, Discover: ai development governance an unauthorized action, missing context, or an interaction the user cannot recover from. The supporting topic, generative system design and controlled outputs, adds this risk: For an owned and testable risk register, Unbounded generation can create unsupported statements, inconsistent formats, sensitive disclosure, or automation that users cannot correct. Each risk management risk needs a detection signal and a response path. The owner of an owned and testable risk register must know when to limit exposure or reopen the decision./pp##Tie mitigation to evidenceThe risk management decision needs evidence that can be revisited. In Building a Useful Delivery Risk Register, End-to-end tests measure task completion, recognition failures, correction paths, tool outcomes, escalation, Overview: generative ai app development services latency, and abandonment. The adjacent topic of generative system design and controlled outputs contributes another requirement. Within risk management, Representative evaluations measure task completion, groundedness, policy behavior, formatting, latency, and escalation outcomes. Store the risk management observation with its owner and date, then keep unresolved limits visible beside the result./p##Carry the result into ownershipThe intended primary outcome is recorded without embellishment: Under Write risks as observable conditions, The interface supports a bounded task and gives users clear ways to confirm, correct, Details: artificial intelligence developing services or leave the automated flow. The supporting outcome for generative system design and controlled outputs is this: For an owned and testable risk register, Users receive a controlled product capability rather than an opaque prompt connected directly to a workflow. Before the next step, an owned and testable risk register should identify scope and exposure; ownership and exit conditions belong in the same record.]]&gt;</description>
      <content:encoded><![CDATA[<p><p>teams building customer and employee assistants often approach AI development services through questions about voice and conversational interaction design. In Building a Useful Delivery Risk Register, A conversational interface must manage recognition errors, interruptions, context, identity, tool calls, Resource: enterprise generative ai development services and user expectations in real time. <a href="https://ai-development-services.com/" rel="nofollow">ai software development Company</a> A risk management brief must resolve which uncertainties require mitigation, acceptance, transfer or a stop decision. For an owned and testable risk register, Discover: What does ai company do? search language such as “conversational ai development services” supplies context for that decision, not evidence that one option is universally suitable.</p><p>#<a href="/alexandriadickey/tag:Turn" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Turn</span></a> related queries into accountable questionsInterest in “generative <a href="https://ai-software-development.net/" title="ai development services company" rel="nofollow">ai development services company</a>”, “<a href="https://ai-software-development.net/" title="ai development services company" rel="nofollow">ai development services company</a> voicebot development services”, “top ai developer companies”, “<a href="https://ai-software-development.net/" title="ai development services company" rel="nofollow">ai development services company</a> voice bot development services”, and “generative ai app development services” creates several entry points to risk management. Reviewers can connect those entry points to explicit limits, observable behavior Explore: ai development consulting and a correction path inside an owned and testable risk register. The resulting owned and Explore: Explore: adaptive ai development services ai proof of concept development services testable risk register record explains what is known, what remains uncertain and which event should reopen the decision.</p><p>#<a href="/alexandriadickey/tag:Write" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Write</span></a> risks as observable conditionsAn owned and testable risk register keeps the risk management discussion reviewable. The source topic states this practice: ai development pros and cons In Building a Useful Delivery Risk Register, Conversation design should define intents, turn handling, confirmation, <a href="https://write.rlp.schule/nicholasleggett9/" title="AI software development company" rel="nofollow">AI software development company</a> repair, escalation, privacy notices, latency, and session state. A connected practice comes from generative system design and controlled outputs: For an owned and testable risk register, Design should separate instruction, context, generation, validation, citation, Details: ai voice agent development services and user correction into observable steps. Together they define what happens before commitment in risk management and what remains in an owned and Read more: generative ai development services testable risk register after the decision.</p><p>#<a href="/alexandriadickey/tag:Set" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Set</span></a> failure boundaries for risk managementThe primary risk record says: For an owned and Discover: ai as a service companies testable risk register, A fluent response can conceal misunderstood input, Discover: ai development governance an unauthorized action, missing context, or an interaction the user cannot recover from. The supporting topic, generative system design and controlled outputs, adds this risk: For an owned and testable risk register, Unbounded generation can create unsupported statements, inconsistent formats, sensitive disclosure, or automation that users cannot correct. Each risk management risk needs a detection signal and a response path. The owner of an owned and testable risk register must know when to limit exposure or reopen the decision.</p><p>#<a href="/alexandriadickey/tag:Tie" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Tie</span></a> mitigation to evidenceThe risk management decision needs evidence that can be revisited. In Building a Useful Delivery Risk Register, End-to-end tests measure task completion, recognition failures, correction paths, tool outcomes, escalation, Overview: generative ai app development services latency, and abandonment. The adjacent topic of generative system design and controlled outputs contributes another requirement. Within risk management, Representative evaluations measure task completion, groundedness, policy behavior, formatting, latency, and escalation outcomes. Store the risk management observation with its owner and date, then keep unresolved limits visible beside the result.</p>#<a href="/alexandriadickey/tag:Carry" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Carry</span></a> the result into ownershipThe intended primary outcome is recorded without embellishment: Under Write risks as observable conditions, The interface supports a bounded task and gives users clear ways to confirm, correct, Details: artificial intelligence developing services or leave the automated flow. The supporting outcome for generative system design and controlled outputs is this: For an owned and testable risk register, Users receive a controlled product capability rather than an opaque prompt connected directly to a workflow. Before the next step, an owned and testable risk register should identify scope and exposure; ownership and exit conditions belong in the same record.</p>
]]></content:encoded>
      <guid>https://blog.0yu.de/alexandriadickey/building-a-useful-delivery-risk-register-ai-development-services</guid>
      <pubDate>Wed, 18 Nov 2026 16:12:24 +0100</pubDate>
    </item>
    <item>
      <title>AI development services: Designing a Pilot That Supports a Decision</title>
      <link>https://blog.0yu.de/alexandriadickey/ai-development-services-designing-a-pilot-that-supports-a-decision</link>
      <description>&lt;![CDATA[pA pilot design review gives AI development services a practical boundary. It connects generative system design and controlled outputs with the needs of teams building text and Read more: ai mobile app development services content features. In Designing a Pilot That Supports a Decision, Generated output must be useful for a real task while remaining bounded by source quality, policy, Overview: ai driven software development services format, and review needs. The governing question is what a limited release must prove before wider investment or exposure. Learn more: ai development services company During pilot design, the query &#34;enterprise generative ai development services&#34; signals the subject a reader wants resolved while acceptance still depends on observed evidence./pp##Connect reader language to the decisionQuestions expressed as &#34;generative ai development services&#34;, &#34;ai voice agent development services&#34;, &#34;what is an ai development company&#34;, &#34;ai chatbot development services&#34;, and &#34;custom generative ai development services&#34; point to adjacent parts of pilot design. The terms help organize discovery, but each one still needs a concrete acceptance condition, an owner and evidence recorded in a pilot protocol with exit criteria. This keeps semantic relevance in a pilot protocol with exit criteria tied to a useful review instead of an unsupported promise./pp##Choose a representative boundaryA pilot protocol with exit criteria keeps the pilot design discussion reviewable. The source topic states this practice: For a pilot protocol with exit criteria, Design should separate instruction, context, generation, validation, citation, and Learn more: ai game development services user correction into observable steps. A connected practice comes from voice and conversational interaction design: In Designing a Pilot That Supports a Decision, Conversation design should define intents, Learn more: Details: ai dev solutions ai mobile app development services turn handling, confirmation, repair, Learn more: ai dating app development services escalation, privacy notices, latency, and session state. Together they define what happens before commitment in pilot design and what remains in a pilot protocol with exit criteria after the decision./pp##Turn uncertainty into a response planUnder Choose a representative boundary, Unbounded generation can create unsupported statements, [empty]&#34;) inconsistent formats, sensitive disclosure, or Explore: ai developer service automation that users cannot correct. That is the first risk considered during pilot design. The second comes from voice and conversational interaction design: Under Choose a representative boundary, A fluent response can conceal misunderstood input, an unauthorized action, missing context, [empty]&#34;) or an interaction the user cannot recover from. A pilot design response plan should pair each trigger with an owner and next action; severity and Explore: ai application development services reversibility can then guide exposure./pp##Define proceed and stop conditionsThe pilot design decision needs evidence that can be revisited. In Designing a Pilot That Supports a Decision, Representative evaluations measure task completion, Read more: custom generative ai development services groundedness, policy behavior, formatting, Discover: ai development services for healthcare latency, and escalation outcomes. The adjacent topic of voice and conversational interaction design contributes another requirement. For a pilot protocol with exit criteria, End-to-end tests measure task completion, recognition failures, correction paths, tool outcomes, escalation, latency, and Details: ai ehr software development services abandonment. Store the pilot design observation with its owner and date, then keep unresolved limits visible beside the result./p##Define what happens after approvalFor generative system design and controlled outputs, the desired operating state is clear: Within pilot design, Users receive a controlled product capability rather than an opaque prompt connected directly to a workflow. The secondary topic adds another state: Under Choose a representative boundary, Read more: multimodal ai development services The interface supports a bounded task and gives users clear ways to confirm, correct, or leave the automated flow. The pilot design record should show how both states will be maintained and Overview: ai powered mvp development services when the decision must be reviewed again.]]&gt;</description>
      <content:encoded><![CDATA[<p><p>A pilot design review gives AI development services a practical boundary. It connects generative system design and controlled outputs with the needs of teams building text and Read more: ai mobile app development services content features. In Designing a Pilot That Supports a Decision, Generated output must be useful for a real task while remaining bounded by source quality, policy, Overview: ai driven software development services format, and review needs. The governing question is what a limited release must prove before wider investment or exposure. <a href="https://ai-software-development.net/" rel="nofollow">Learn more: ai development services company</a> During pilot design, the query “enterprise generative ai development services” signals the subject a reader wants resolved while acceptance still depends on observed evidence.</p><p>#<a href="/alexandriadickey/tag:Connect" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Connect</span></a> reader language to the decisionQuestions expressed as “generative ai development services”, “ai voice agent development services”, “what is an ai development company”, “ai chatbot development services”, and “custom generative <a href="https://ai-development-services.com/" title="ai development services" rel="nofollow">ai development services</a>” point to adjacent parts of pilot design. The terms help organize discovery, but each one still needs a concrete acceptance condition, an owner and evidence recorded in a pilot protocol with exit criteria. This keeps semantic relevance in a pilot protocol with exit criteria tied to a useful review instead of an unsupported promise.</p><p>#<a href="/alexandriadickey/tag:Choose" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Choose</span></a> a representative boundaryA pilot protocol with exit criteria keeps the pilot design discussion reviewable. The source topic states this practice: For a pilot protocol with exit criteria, Design should separate instruction, context, generation, validation, citation, and Learn more: ai game development services user correction into observable steps. A connected practice comes from voice and conversational interaction design: In Designing a Pilot That Supports a Decision, Conversation design should define intents, Learn more: Details: ai dev solutions ai mobile app development services turn handling, confirmation, repair, Learn more: ai dating app development services escalation, privacy notices, latency, and session state. Together they define what happens before commitment in pilot design and what remains in a pilot protocol with exit criteria after the decision.</p><p>#<a href="/alexandriadickey/tag:Turn" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Turn</span></a> uncertainty into a response planUnder Choose a representative boundary, Unbounded generation can create unsupported statements, <a href="https://image.google.nr/url?q=https://ai-development-services.com/" title="[empty]" rel="nofollow">[empty]</a> inconsistent formats, sensitive disclosure, or Explore: ai developer service automation that users cannot correct. That is the first risk considered during pilot design. The second comes from voice and conversational interaction design: Under Choose a representative boundary, A fluent response can conceal misunderstood input, an unauthorized action, missing context, <a href="https://bio.rocketapps.pro/louisaloy7" title="[empty]" rel="nofollow">[empty]</a> or an interaction the user cannot recover from. A pilot design response plan should pair each trigger with an owner and next action; severity and Explore: ai application development services reversibility can then guide exposure.</p><p>#<a href="/alexandriadickey/tag:Define" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Define</span></a> proceed and stop conditionsThe pilot design decision needs evidence that can be revisited. In Designing a Pilot That Supports a Decision, Representative evaluations measure task completion, <a href="http://ingeekswetrust.de/index.php?title=Aligning_Stakeholders_Around_One_Delivery_Contract:_AI_Development_Services" rel="nofollow">Read more: custom generative ai development services</a> groundedness, policy behavior, formatting, Discover: ai development services for healthcare latency, and escalation outcomes. The adjacent topic of voice and conversational interaction design contributes another requirement. For a pilot protocol with exit criteria, End-to-end tests measure task completion, recognition failures, correction paths, tool outcomes, escalation, latency, and Details: ai ehr software development services abandonment. Store the pilot design observation with its owner and date, then keep unresolved limits visible beside the result.</p>#<a href="/alexandriadickey/tag:Define" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Define</span></a> what happens after approvalFor generative system design and controlled outputs, the desired operating state is clear: Within pilot design, Users receive a controlled product capability rather than an opaque prompt connected directly to a workflow. The secondary topic adds another state: Under Choose a representative boundary, Read more: multimodal <a href="https://ai-development-services.com/" title="ai development services" rel="nofollow">ai development services</a> The interface supports a bounded task and gives users clear ways to confirm, correct, or leave the automated flow. The pilot design record should show how both states will be maintained and Overview: ai powered mvp development services when the decision must be reviewed again.</p>
]]></content:encoded>
      <guid>https://blog.0yu.de/alexandriadickey/ai-development-services-designing-a-pilot-that-supports-a-decision</guid>
      <pubDate>Wed, 18 Nov 2026 14:44:55 +0100</pubDate>
    </item>
    <item>
      <title>Assessing Data Readiness for Delivery: AI development services</title>
      <link>https://blog.0yu.de/alexandriadickey/assessing-data-readiness-for-delivery-ai-development-services</link>
      <description>&lt;![CDATA[pdata owners, Explore: ai development services for startups architects, and Discover: ai development pros and cons product teams often approach AI development services through questions about data readiness and information contracts. Read more: ai agent development services Within data readiness, A promising use case may depend on information that is incomplete, inaccessible, poorly governed, or unavailable at decision time. A data readiness brief must resolve whether the product can obtain and govern the information required at decision time. For Guide: Guide: Resource: ai dev solutions ai development services for healthcare ai development services company a data readiness inventory, search language such as &#34;ai ml software development services&#34; supplies context for that decision, not evidence that one option is universally suitable./pp##Connect reader language to the decisionQuestions expressed as &#34;ai proof of concept development services&#34;, &#34;what does ai company do&#34;, &#34;what is ai development framework&#34;, and &#34;ai software development services&#34; point to adjacent parts of data readiness. The terms help organize discovery, Discover: Read more: ai full stack development services ai full stack development services but each one still needs a concrete acceptance condition, an owner and ai healthcare software development services evidence recorded in a data readiness inventory. This keeps semantic relevance in a data readiness inventory tied to a useful review instead of an unsupported promise./pp##Trace information to its ownerA data readiness inventory keeps the data readiness discussion reviewable. The source topic states this practice: For Learn more: How to build ai service? a data readiness inventory, Teams should define sources, Explore: ai development firm ownership, freshness, permissions, quality checks, retention, and fallback behavior before model integration. A connected practice comes from proof of concept and minimum viable product planning: Within data readiness, A bounded experiment should name the hypothesis, representative inputs, baseline, evaluation method, time box, ai powered software development services and Resource: ai as a service companies stop condition. Together they define what happens before commitment in data readiness and what remains in a data readiness inventory after the decision./pp##Turn uncertainty into a response planWithin data readiness, Hidden data assumptions can produce unreliable behavior, Details: ai recommendation engine development services privacy exposure, delayed delivery, or [empty]&#34;) a system that cannot be operated legally. That is the first risk considered during data readiness. The second comes from proof of concept and minimum viable product planning: Within data readiness, Learn more: What is an ai development company? A prototype can appear successful while avoiding integration, security, latency, [empty]&#34;) failure handling, and ai voice bot development services maintenance constraints. A data readiness response plan should pair each trigger with an owner and next action; severity and reversibility can then guide exposure./pp##Plan for missing and changing dataEvidence attached to a data readiness inventory should retain the primary topic&#39;s rule: In Assessing Data Readiness for Delivery, A data contract records fields, provenance, access controls, expected quality, update behavior, and test fixtures for representative cases. The supporting evidence for proof of concept and what is ai development framework minimum viable product planning is also explicit: Guide: fintech ai development services In Assessing Data Readiness for Delivery, The experiment record should show tested cases, observed limitations, unresolved risks, and the decision supported by the result. A data readiness inventory identifies its source and Discover: ai game development services version; it also preserves exceptions and Details: ai powered development services the next decision./p##Carry the result into ownershipThe intended primary outcome is recorded without embellishment: Under Trace information to its owner, Implementation decisions are grounded in information the product can actually obtain and maintain. The supporting outcome for proof of concept and minimum viable product planning is this: Within data readiness, The organization gains evidence for a proceed, revise, buy, or stop decision without inheriting an accidental production system. Before the next step, a data readiness inventory should identify scope and exposure; ownership and exit conditions belong in the same record.]]&gt;</description>
      <content:encoded><![CDATA[<p><p>data owners, Explore: <a href="https://ai-development-services.com/" title="ai development services" rel="nofollow">ai development services</a> for startups architects, and Discover: ai development pros and cons product teams often approach <a href="https://ai-development-services.com/" title="AI development services" rel="nofollow">AI development services</a> through questions about data readiness and information contracts. <a href="https://ai-software-development.net/" rel="nofollow">Read more: ai agent development services</a> Within data readiness, A promising use case may depend on information that is incomplete, inaccessible, poorly governed, or unavailable at decision time. A data readiness brief must resolve whether the product can obtain and govern the information required at decision time. For Guide: Guide: Resource: ai dev solutions ai development services for healthcare <a href="https://ai-development-services.com/" title="ai development services" rel="nofollow">ai development services</a> company a data readiness inventory, search language such as “ai ml software development services” supplies context for that decision, not evidence that one option is universally suitable.</p><p>#<a href="/alexandriadickey/tag:Connect" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Connect</span></a> reader language to the decisionQuestions expressed as “ai proof of concept development services”, “what does ai company do”, “what is ai development framework”, and “ai software development services” point to adjacent parts of data readiness. The terms help organize discovery, Discover: Read more: ai full stack development services ai full stack development services but each one still needs a concrete acceptance condition, an owner and ai healthcare software development services evidence recorded in a data readiness inventory. This keeps semantic relevance in a data readiness inventory tied to a useful review instead of an unsupported promise.</p><p>#<a href="/alexandriadickey/tag:Trace" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Trace</span></a> information to its ownerA data readiness inventory keeps the data readiness discussion reviewable. The source topic states this practice: For Learn more: How to build ai service? a data readiness inventory, Teams should define sources, Explore: ai development firm ownership, freshness, permissions, quality checks, retention, and fallback behavior before model integration. A connected practice comes from proof of concept and minimum viable product planning: Within data readiness, A bounded experiment should name the hypothesis, representative inputs, baseline, evaluation method, time box, ai powered software development services and Resource: ai as a service companies stop condition. Together they define what happens before commitment in data readiness and what remains in a data readiness inventory after the decision.</p><p>#<a href="/alexandriadickey/tag:Turn" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Turn</span></a> uncertainty into a response planWithin data readiness, Hidden data assumptions can produce unreliable behavior, Details: ai recommendation engine development services privacy exposure, delayed delivery, or <a href="https://image.google.to/url?q=https://ai-development-services.com/" title="[empty]" rel="nofollow">[empty]</a> a system that cannot be operated legally. That is the first risk considered during data readiness. The second comes from proof of concept and minimum viable product planning: Within data readiness, Learn more: What is an ai development company? A prototype can appear successful while avoiding integration, security, latency, <a href="https://toolbarqueries.google.com.ec/url?q=https://ai-software-development.net/" title="[empty]" rel="nofollow">[empty]</a> failure handling, and ai voice bot development services maintenance constraints. A data readiness response plan should pair each trigger with an owner and next action; severity and reversibility can then guide exposure.</p><p>#<a href="/alexandriadickey/tag:Plan" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Plan</span></a> for missing and changing dataEvidence attached to a data readiness inventory should retain the primary topic&#39;s rule: In Assessing Data Readiness for Delivery, A data contract records fields, provenance, access controls, expected quality, update behavior, and test fixtures for representative cases. The supporting evidence for proof of concept and what is ai development framework minimum viable product planning is also explicit: Guide: fintech ai development services In Assessing Data Readiness for Delivery, The experiment record should show tested cases, observed limitations, unresolved risks, and the decision supported by the result. A data readiness inventory identifies its source and Discover: ai game development services version; it also preserves exceptions and Details: ai powered development services the next decision.</p>#<a href="/alexandriadickey/tag:Carry" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">Carry</span></a> the result into ownershipThe intended primary outcome is recorded without embellishment: Under Trace information to its owner, Implementation decisions are grounded in information the product can actually obtain and maintain. The supporting outcome for proof of concept and minimum viable product planning is this: Within data readiness, The organization gains evidence for a proceed, revise, buy, or stop decision without inheriting an accidental production system. Before the next step, a data readiness inventory should identify scope and exposure; ownership and exit conditions belong in the same record.</p>
]]></content:encoded>
      <guid>https://blog.0yu.de/alexandriadickey/assessing-data-readiness-for-delivery-ai-development-services</guid>
      <pubDate>Wed, 18 Nov 2026 12:30:49 +0100</pubDate>
    </item>
  </channel>
</rss>