(maka) branch main updated: docs(blog): add conclusion to Log Is the Runtime (#4635)

[ku...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

likun666661 pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/maka.git


The following commit(s) were added to refs/heads/main by this push:
     new 079f784a59 docs(blog): add conclusion to Log Is the Runtime (#4635)
079f784a59 is described below

commit 079f784a59b2717b46460458eca754f7fe9c5386
Author: KunLi <90...@users.noreply.github.com>
AuthorDate: Thu Sep 3 16:05:40 2026 +0800

    docs(blog): add conclusion to Log Is the Runtime (#4635)
    
    Generated-by: OpenAI Codex
    
    Co-authored-by: likun <ku...@alva.xyz>
---
 docs/blogs/log-is-the-runtime.md       | 12 ++++++++++++
 docs/blogs/log-is-the-runtime.zh-CN.md | 12 ++++++++++++
 2 files changed, 24 insertions(+)

diff --git a/docs/blogs/log-is-the-runtime.md b/docs/blogs/log-is-the-runtime.md
index cb4ace916a..67c8fd315b 100644
--- a/docs/blogs/log-is-the-runtime.md
+++ b/docs/blogs/log-is-the-runtime.md
@@ -203,3 +203,15 @@ Maka applies offloading across two operational horizons:
 Pruning affects only the projected view visible to the model. The canonical `RuntimeEvent Log` retains original tool outputs, ensuring subsequent history compaction summarizes real operational facts rather than placeholder markers.
 
 History Compaction compresses along the temporal axis, folding older sequential turns into continuous semantic summaries. Tool Result Pruning offloads along the payload axis, moving oversized results into backing storage while preserving fine-grained event structure. Together, they keep inference fast and lean without sacrificing execution truth.
+
+## Conclusion: Preserve Facts, Defer Representation
+
+Maka's design follows a consistent rule: record what happened reliably, then decide how each consumer should read it.
+
+The UI, model context, crash recovery, and task continuation all read from the same `RuntimeEvent Log`, but each uses a different projection. Compaction changes the resolution at which history is represented. Tool Result Pruning changes how large payloads enter the context. Neither requires rewriting facts that have already been committed.
+
+Append-only history does not remove complexity. It moves complexity away from maintaining a mutable current state and into constructing appropriate views over stable history. The trade-off is continued log growth, versioned and verified projections, and lifecycle management for archived data. In return, the runtime gains clearer recovery boundaries, a more complete audit trail, and the ability to reinterpret history as models and context policies evolve.
+
+For an agent runtime, state management is not about keeping the entire history inside the model forever. It is about preserving a complete, verifiable factual record and constructing a bounded, useful context before each inference step.
+
+The model decides what to do next. The log ensures that the runtime can always reconstruct what has already happened. This is the practical meaning of **Log Is the Runtime**.
diff --git a/docs/blogs/log-is-the-runtime.zh-CN.md b/docs/blogs/log-is-the-runtime.zh-CN.md
index 2abe09f87b..e5936c57b8 100644
--- a/docs/blogs/log-is-the-runtime.zh-CN.md
+++ b/docs/blogs/log-is-the-runtime.zh-CN.md
@@ -200,3 +200,15 @@ Maka 借鉴了操作系统的按需分页（Demand Paging）思想：在把完
 这两类裁剪仅作用于对模型可见的 Projection。原始的 `RuntimeEvent Log` 中永远保留完整的工具输出，后续的 History Compaction 在生成语义摘要时看到的也是真实事件，而不是占位符。
 
 History Compaction 沿时间轴压缩历史，将长事件序列折叠为低分辨率的语义摘要；Tool Result Prune 沿空间轴卸载负载，保持事件结构的同时将大对象转移到持久存储中。两者以不可变的 Append-Only Log 为基准，在减少 Token 占用的同时，保证了执行历史的精确与可复原。
+
+## 结语：保留事实，延迟决定如何读取
+
+Maka 的这些设计最终指向同一个原则：先把发生过的事情可靠地记录下来，再根据不同场景决定如何读取它们。
+
+UI、模型上下文、崩溃恢复和任务续跑读取的是同一份 `RuntimeEvent Log`，但各自使用不同的 Projection。Compaction 改变历史的表达粒度，Tool Result Prune 改变大对象进入上下文的方式，它们都不需要改写已经发生的事实。
+
+Append-Only 并没有消除系统复杂度，而是把复杂度从维护一份不断被覆盖的当前状态，转移到了如何从稳定历史构造合适的视图。这样做的代价是日志会持续增长，Projection 需要版本和校验，归档数据也必须管理生命周期。但它换来的是更清晰的恢复边界、更完整的审计能力，以及在模型和上下文策略变化后重新解释历史的可能性。
+
+对于 Agent Runtime 来说，状态管理的目标并不是把全部历史永远塞进模型，而是保留一份完整、可验证的事实记录，并在每次推理前构造一份有界且有用的上下文。
+
+模型决定下一步做什么，Log 则保证 Runtime 始终能够回答：此前究竟发生过什么。这就是 **Log Is the Runtime**。


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@maka.apache.org
For additional commands, e-mail: commits-help@maka.apache.org